WO1997003525A1 - Grid moving method of object image and apparatus using the same and compaction/motion estimation method using the same and apparatus thereof - Google Patents

Grid moving method of object image and apparatus using the same and compaction/motion estimation method using the same and apparatus thereof Download PDF

Info

Publication number
WO1997003525A1
WO1997003525A1 PCT/KR1996/000106 KR9600106W WO9703525A1 WO 1997003525 A1 WO1997003525 A1 WO 1997003525A1 KR 9600106 W KR9600106 W KR 9600106W WO 9703525 A1 WO9703525 A1 WO 9703525A1
Authority
WO
WIPO (PCT)
Prior art keywords
grid
axis
image
moving
counting
Prior art date
Application number
PCT/KR1996/000106
Other languages
French (fr)
Inventor
Sung-Moon Chun
Gwang-Hoon Park
Joo-Hee Moon
Jae-Kyoon Kim
Kuk-Yeol Yoo
Original Assignee
Hyundai Electronics Industries Co., Ltd.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from KR1019950037919A external-priority patent/KR970025172A/en
Priority claimed from KR1019960006465A external-priority patent/KR970009394A/en
Application filed by Hyundai Electronics Industries Co., Ltd. filed Critical Hyundai Electronics Industries Co., Ltd.
Priority to DE1996615805 priority Critical patent/DE69615805T2/en
Priority to AU63199/96A priority patent/AU6319996A/en
Priority to EP96922276A priority patent/EP0781492B1/en
Priority to CA 2198463 priority patent/CA2198463C/en
Publication of WO1997003525A1 publication Critical patent/WO1997003525A1/en

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/60Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding
    • H04N19/649Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding the transform being applied to non rectangular image segments
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/102Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
    • H04N19/119Adaptive subdivision aspects, e.g. subdivision of a picture into rectangular or non-rectangular coding blocks
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/50Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
    • H04N19/503Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving temporal prediction
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/50Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
    • H04N19/503Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving temporal prediction
    • H04N19/51Motion estimation or motion compensation
    • H04N19/527Global motion vector estimation
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/50Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding
    • H04N19/503Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using predictive coding involving temporal prediction
    • H04N19/51Motion estimation or motion compensation
    • H04N19/537Motion estimation other than block-based
    • H04N19/54Motion estimation other than block-based using feature points or meshes
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/85Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using pre-processing or post-processing specially adapted for video compression
    • H04N19/86Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using pre-processing or post-processing specially adapted for video compression involving reduction of coding artifacts, e.g. of blockiness
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/20Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using video object coding

Definitions

  • the present invention relates to a grid moving method for inimizing image information of an object image and an apparatus therefore and a compaction/motion estimation method using the grid moving method and an apparatus therefor, and particularly to an improved grid moving method for an object image and an apparatus therefor and a compaction/motion estimation method using the grid moving method and an apparatus therefor which are capable of forming a grid with respect to an image of a predetermined object having shape information and.in a region of an image, dividing the region of the image into a plurality of unit regions, moving the formed grid, and detecting a position at which the amount of information is reduced when performing a compaction or estimating a motion of an object.
  • the present invention is basically directed to reforming the grid to a position at which the amount of information is reduced by moving the grid, and is directed to separating and coding each unit region in which an image of the object exist from the reformed grid and using the grid movement of the image of an object of which the motion is estimated.
  • the image of the object is coded, and the motion of the image is estimated and then the amount of the information in the image is reduced for storing the information in a predetermined recording/writing medium. Thereafter, the information is transmitted to a predetermined destination in real time.
  • a vector quantum ization method or a discrete cosine transform (DCT) method is used.
  • SADCT shape adaptive discrete cosine transform
  • the above-mentioned shape adaptive discrete cosine transform method is directed to forming a grid with respect to an image frame, dividing an image of an object into a plurality of unit regions each having predetermined size and shape information, separating a unit region from the plurality of the unit regions in which an image of the object exists and then coding the unit region.
  • the unit region contains an image to be coded
  • the effectiveness between a two-dimensional region DCT and a compaction become identical in the shape adaptive discrete cosine transform.
  • the pixel in which an image of an object exists, is processed with respect to the X-axis in a one-dimensional discrete cosine transform method, and a result of the above X-axis-based process is processed with respect to the Y-axis in a one-dimensional discrete cosine transform method. Thereafter, the final result value is obtained.
  • the shape adaptive discrete cosine transform method is further directed to reducing the number of unit regions in which an image of an object exists and performing the compaction after substantially filling the image of the object in the unit region, thus enhancing the compaction of a transform constant.
  • the image of the object to be coded should preferably be substantially filled in each unit region, and then the number of unit regions in which an image of the object exists is effectively reduced.
  • Figs. 1A and 1B show grid patterns formed in one frame.
  • one frame is divided into a plurality of rows and columns which are consisted of a plurality of unit regions 21 having the same size and shape in cooperation with a P X Q number of X-axis grid and Y-axis grid 11 and 13 spaced apart from one another at a regular distance.
  • a unit region 21 may be formed in various shapes.
  • the unit region 21 is formed in a regular square or a rectangular form by the X-axis and Y-axis grid 11 and 13.
  • a unit region 21 may be formed as a horizontally lying triangle or a horizontally upside down triangle, and neighboring triangles form rectangular shapes bounded by the slant grids 15 and 17.
  • a unit region 21 is formed by vertically lying triangles and neighboring triangles form rectangular shapes bounded by the slant grids 15 and 17.
  • the unit region 21 is formed as a 45° rotated square by the slant grids 15 and 17, and as shown in Figs. 2D and 2E, the unit region 21 is foimed in a hexagonal shape by the slant grids 15 and 17. As shown in Fig. 2F, the unit region 21 is formed in an octagonal form having a 45° rotated small square between the neighboring octagons. In this example, two different shaped unit regions 21 are concurrently used. Any shape which spatially and evenly divides the image frame may be used for the unit region 21.
  • a square- or rectangular-shaped unit region 21 which is defined by an X-axis grid 11 and Y-axis grid 13 will now be explained.
  • the unit region 21 is formed of an M X N number of unit pixels 23 in the X-axis and Y-axis directions.
  • one unit region 21 is formed of an 8 X 8 number of unit pixels 23 or is formed of a 16 X 16 number of unit pixels 23.
  • a unit region 21 is defined as an M X N number of blocks in accordance with the number of unit pixels 23. As shown in Fig. 1B, the unit region 21 refers to an 8 X 8 number of blocks corresponding to unit pixels.
  • Fig. 3A shows an image (shown as the hatched portion) having predetermined shape information in a unit region 21 formed of an 8 X 8 number of unit pixels 23.
  • the shape adaptive discrete cosine transform with respect to the image of an object as shown in Fig. 3B, the image of the object is filled from the upper side margin portion of the unit region 21, and then the one-dimensional cosine transform is performed with respect to the Y-axis which is shown in the vertical direction.
  • the one-dimensional discrete cosine transform is performed as shown in Fig. 3D.
  • the image of the object is filled from the left side margin portion of the unit region 21, as shown in Fig. 3E,and then the one-dimensional discrete cosine transform is performed with respect to the X-axis which is shown in the horizontal direction.
  • the one-dimensional discrete cosine transform is completed with respect to the X-axis, as shown in Fig. 3F, the shape adaptive discrete cosine transform with to the Y-axis and X-axis is completed.
  • a zig-zag scan is performed with respect to the final shape, as shown in Fig. 3F, which is obtained by the above-mentioned shape adaptive discrete cosine transform.
  • the zig-zag scan is performed diagonally from the leftmost side and the uppermost side to the rightmost side and the lowermost side.
  • the conventional shape adaptive discrete cosine transform is directed to performing the shape adaptive discrete cosine transform in accordance with the position in which the image of an object exists without moving the position of the grid.
  • the bit rate per frame is high, and since the number of the unit regions in which the image of the object exists is numerous, there is a restriction on the ability to reduce an amount of compaction information which is obtained by coding the image of the object and the amount of motion information which is obtained by estimating the motion of the object.
  • an object-based moving image coding method is generally used in the industry.
  • the above-mentioned object-based moving image coding method is directed to segmenting the image of the object in a background in which there is not a moving image and a changed region which is defined by the moving image of the object.
  • the moving object of the changed region is segmented into a motion compensable object and a motion compensable failed object through motion estimation.
  • the motion compensable object refers to the moving object having a predetermined theory such as a horizontal movement, a rotational movement, a lineal movement, and the like in a state that the object in a three-dimensional space is converted into a two-dimensional image of the object.
  • the motion compensable failed object refers to an object which is not adaptable with respect to the above-mentioned theory.
  • the motion compensable object process is directed to detecting motion information of the image of the object.
  • the image of the motion compensable failed object and the image of the exposed object are most effectively coded so as to reduce the amount of information, which is then transmitted and stored. Since the amount of information with respect to the image of the motion compensable failed object is about 60-70% of the total amount of the information to be transmitted, many studies have been conducted, in the industry, so as to reduce the amount of information transmitted.
  • the motion estimation of the motion compensable object is directed to segmenting and estimating the moving portion of the moving image from a picture of the previous frame so as to minimize the amount of motion information.
  • the present invention is basically directed to segmenting an image of an object having predetermined shape information with a grid, detecting a position in which the amount of this information can be reduced by moving the position of the grid along the X-axis or Y-axis directions, and moving the grid to the position in which the amount of information can be reduced.
  • the present invention is directed to moving the position of the grid so that the image of the object can be positioned in the minimum number of unit regions so as to reduce the amount of information with respect to the image of the object.
  • the present invention is directed to detecting the number of unit regions in which the image of an object exists and moving the position of the grid so that the image of the object exists in the minimum number of unit regions and coding the image of the object existing in the detected unit regions so as to minimize the amount of information.
  • the present invention is directed to judging the unit regions in which the image of the object exists in a state that the position of the grid is moved so that the image of the object can exists in the minimum number of unit regions, and estimating the motion of the object using the judged unit region and detecting the position of the grid using the information of the object.
  • a grid moving method of an object image including the steps of: a segmenting step for forming a grid over an image of an object having predetermined shape information and for segmenting the image into a plurality of unit regions; and a detecting step for detecting a position at which the amount of information is reduced by moving the grid formed in the segmenting step.
  • a compaction/motion estimation method including the steps of: a moving step for forming a grid over an image of an object having predetermined shape information, segmenting the image into a plurality of unit regions, and moving the foimed grid; a judging step for judging the amount of information at each posion to which the grid is moved in the moving step; a detecting step for detecting a position at which the amount of information is reduced; a compaction step for reforming the grid in accordance with the position detected in the detecting step and for coding the image of the object existing in each unit region of the reformed grid; and a motion estimation step for reforming the grid in accordance with the position detected in the detecting step and for estimating the motion of the image of the object in the unit regions in which the image of the object exists among the unit regions segmented by the reformed grid.
  • a compaction method including the steps of: a moving step for forming a grid over an image of an object having predetermined shape information, segmenting the image into a plurality of unit regions, and for moving the formed grid; a judging step for judging the amount of information at each position to which the grid is moved in the moving step; a detecting step for detecting a position at which the amount of information is reduced in the judging step; and a compaction step for reforming the grid in accordance with the position detected in the detecting step and coding the image of the object existing in the unit region of the reformed grid.
  • a motion estimation method including the steps of: a moving step for forming a grid over an image of an object having predetermined shape information, segmenting the image into a plurality of unit regions, and moving the formed grid; a judging step for judging the amount of information at each position to which the grid is moved in the moving step; a detecting step for detecting a position at which the amount of information is reduced in the judging step; and a motion estimation step for reforming the grid in accordance with the position detected in the detecting step and for estimating the motion of the image of the object in the unit regions in which the image of the object exist among the unit regions by the reformed grid.
  • a compaction/motion estimation method including the steps of: a separating step for estimating the motion of an image of an object having predetermined shape information and for separating a motion compensable failed object image and a motion compensable object image; a first moving step for forming a grid over the motion compensable failed object image separated in the separating step, segmenting the image into a plurality of unit regions, and moving the grid; a first judging step for judging the amount of information at each position to which the grid is moved in the first moving step; a first detecting step for detecting a position at which the amount of information is reduced in the first judging step; a compaction step for reforming the grid in accordance with the position detected in the first detecting step and for coding the unit regions in which the image of the motion compensable failed object exists from among the unit regions which is segmented by the reformed grid; a second moving step for forming a grid over the image of the motion compensable object separated in the separating step, segmenting the image into a
  • a compaction method using a grid moving method for an object image including the steps of: a separating step for estimating the motion of the image of an object having shape information and for separating the image of a motion compensable failed object; a moving step for forming a grid over the image of the motion compensable failed object separated in the separating step, segmenting the image into a plurality of unit regions, and moving the grid; a judging step for judging an amount of information at each position to which the grid is moved in the moving step; a detecting step for detecting a position at which the amount of information is reduced in the judging step; and a compaction step for reforming the grid in accordance with the position detected in the etecting step and for coding the unit regions in which the image of the motion compensable failed object exists among the unit regions segmented by the reformed grid.
  • a motion estimation method using a grid movement of an image of an object including the steps of: a separating step for estimating the motion of the image of the object having shape information and for separating the image of the motion compensable object; a moving step for forming a grid over the image of the motion compensable object separated in the separating step, segmenting the image into a plurality of unit regions, and moving the grid; a judging step for judging an amount of information at the position to which the grid is moved in the moving step; a detecting step for detecting a position at which the amount of information is reduced in the judging step; and a motion estimating step for reforming a grid in accordance with the position detected in the detecting step and for estimating the motion of the image of the motion compensable object using the unit regions in which the image of the motion compensable object exists from among the unit regions segmented by the reformed grid.
  • a grid moving apparatus for an object image, including: an address generation controller for moving an address start position at which an address is generated by a predetermined distance within a predetermined region of a unit region; an address generator for separating the image of the object into unit regions in accordance with the address start position which the address generation controller outputs and for generating the address; a memory unit for storing the image of the object having inputted shape information and for outputting the image in accordance with an address outputted from the address generator; a region number counter for counting the number of unit regions in which the shape information of the object exists outputted from the memory unit; and minimum unit region grid selector for selecting an X-axis grid start position XM and a Y-axis grid start position YN at which the minimum number of unit regions is counted from among the number of the unit regions counted by the region number counter.
  • a motion estimation apparatus using a grid moving of an object image including: a grid moving unit for adjusting the grid in accordance with an image position of an object having shape information outputted from an image signal input unit and for reducing the number of unit regions in which the image of the object exists; and a motion estimation unit for estimating a motion of the image of object using the motion of the unit regions of which the number of unit regions is reduced.
  • a compaction/motion estimation apparatus using a grid moving of an object image, including: an image separator for separating an image of a motion compensable failed object and an image of a motion compensable object in accordance with a motion from an image of an object having shape information; a first grid moving unit for adjusting the grid in accordance with an image position of the motion compensable failed object separated by the image separating unit and for reducing the number of unit regions in which the image of the motion compensable failed object exists; a compaction unit for coding the image of the object existing in the unit regions of which the number of the unit regions is reduced by the first grid moving means; a second grid moving unit for adjusting the grid in accordance with the image position of the motion compensable object separated by the image separating unit and for reducing the number of unit regions in which the image of the motion compensable object exists; and a motion estimation unit for estimating the motion information of the motion compensable object using the unit regions in which the image of the motion compensable object exists of which the number of the unit regions
  • a motion estimation apparatus using a grid moving of an object image including: an image separating unit for separating an image of a motion compensable object in accordance with a motion of an object having shape information; a grid moving unit for adjusting a grid in accordance with an image position of the motion compensable object separated by the image separating unit and for reducing the number of unit regions in which the image of the motion compensable object exists; and a motion estimation unit for estimating motion information of the motion compensable object using the unit region in which the image of the motion compensable object exists of which the number of the unit regions is reduced by the grid moving unit.
  • Figs. 1A and 1B are views showing a conventional grid pattern formed in one image frame, of which: Fig. 1A is a view showing a conventional grid pattern formed in one image frame; and
  • Fig. 1B is a view showing a conventional unit region which is indicated as 8 X 8 pixels;
  • Figs. 2A through 2F are views showing various shapes of a conventional unit region;
  • Figs. 3A through 3F are views showing a conventional shape adaptive discrete cosine transform process
  • Fig. 4 a block diagram showing a grid moving apparatus according to the present invention
  • Fig. 5 is a block diagram showing an address generator of Fig. 4 according to the present invention.
  • Fig. 6 is a view showing an order that an address generator outputs an image of an object stored in a memory of Fig. 4 by separating in accordance with an X-axis address and a Y-axis address according to the present invention
  • Fig. 7 is a block diagram showing a region number counter of Fig. 4 according to the present invention.
  • Fig. 8 is a flow chart showing a method for detecting a position at which the amount of information is reduced in a grid moving method according to a first embodiment of the present invention
  • Figs. 9A through 9F are views showing a method for extracting an image of an object in a grid moving method and detecting a position at which the amount of information existing in a unit region is a minimum number according to the present invention
  • Fig. 10 is a flow chart showing another method for detecting a position at which the amount of information is reduced in a grid moving method according to a second embodiment of the present invention.
  • Fig. 11 is a flow chart showing another method for detecting a position at which the amount of information is reduced in a grid moving method according to a third embodiment of the present invention
  • Figs. 12A and 12B are views showing a method for sequentially outputting an image of a unit region by moving an X-axis and Y-axis grid start positions in a zig-zag manner in accordance with a signal flow of Fig. 11 according to the third embodiment of the present invention
  • Fig. 13 is a view showing another method for detecting a position at which the amount of information is reduced in a grid moving method according to a fourth embodiment of the present invention.
  • Fig. 14 is a view showing another method for detecting a position at which the amount of information is reduced in a grid moving method according to a fifth embodiment of the present invention
  • Figs. 15 and 16 show tables showing a comparative result value of the conventional method and of the methods of the various embodiments accrding to the present invention, of which:
  • Fig. 15 is a table showing a comparative result value after moving a grid start potion in accordance with an image position of a young woman.
  • Fig. 16 is a table showing a comparative result value after moving a grid in accordance with an image position of an old woman
  • Fig. 17 is a block diagram showing a motion estimation/compaction apparatus according to the first embodiment of the present invention
  • Figs. 18A through 18C are views showing an image which is reformed by detecting a position at which the amount of information is reduced from an image of a moving compensable object in a motion estimation method according to the present invention
  • Fig. 19 is a block diagram showing another motion estimation/compaction apparatus according to the second embodiment of the present invention.
  • Fig. 20 is a view showing an image frame so as to explain a time lapse-based variation according to the present invention.
  • a grid moving method for minimizing image information of an object and an apparatus using the grid moving method and a compaction/motion estimation method using the grid moving method and an apparatus therefor will now be described with reference to Figs. 4 through 20.
  • reference numeral 31 denotes an address generation controller for moving an address start position at which an address is generated within the range of the X-axis and Y-axis by a predetermined distance
  • reference numeral 33 denotes an address generator for generating X-axis and Y-axis addresses so that the image of an object can be segmented to a unit region and sequentially outputted in accordance with an address start position which the address generation controller 31 generates
  • reference numeral 35 denotes a memory for sequentially storing an image of an object having predetermined shape information and segmenting and outputting a unit region in accordance with an address which the address generator 33 generates. As shown in Fig.
  • an X-axis range determination unit 331 determines the range of the X-axis of a unit region in accordance with the size information of the inputted image
  • the Y-axis range determination unit 333 determines the range of the Y-axis of a unit region in accordance with the size information of the inputted image.
  • the size of the unit region can be determined using either the range of the X-axis or the range of the Y-axis.
  • the region address generator 335 of the address generator 33 judges the ranges of the X-axis and Y-axis of the unit region which is determined by the X-axis range determination unit 331 and the Y-axis range determination unit 333, segments the X-axis and Y-axis with respect to the unit region of the image of the object stored in the memory 35 based on the address start position outputted from the address generation controller 31 , and sequentially outputs the X-axis and Y-axis addresses of the segmented unit region.
  • the memory 35 separates the image stored therein by one unit region from another unit region in accordance with X-axis and Y-xis addresses which are sequentially generated by the region address generator 335. For example, as shown in Fig. 6, the image of the object of the unit region is sequentially outputted, and then the image of the object of the next unit region is sequentially outputted.
  • reference numeral 37 denotes a region number counter for counting the number of unit regions in which the image of the object among the signals outputted from the memory 35 exists.
  • a unit region counter 371 counts the clock signal and segments the unit region.
  • a judging unit 373 separates the unit region of the outputted object in accordance with the output signal of the unit region counter 371 and judges as to whether the image of the object exists.
  • An adder 375 adds the judgment signal of the judging unit 373 and then outputs the number of unit regions in which the image of the object exists.
  • reference numeral 39 denotes a minimum unit region grid selector for selecting the X-axis and Y-axis grid positions at which a minimum number of the regions is counted by the region number counter 37 .
  • the minimum unit region grid selector 39 stores the count value when the counting of region number counter 37 is completed, and controls the address generation controller 31 to move the start positions of the X-axis and Y-axis addresses within the range of the X-axis and Y-axis of the unit region by a predetermined distance.
  • the minimum unit region grid selector 39 moves the start positions of the X-axis and Y-axis addresses and controls the address generation controller 31 when the counting of the number of the unit regions is completed, and then moves the start positions of the X-axis and Y-axis addresses. The above routines are repeated.
  • the minimum unit region grid selector 39 judges the start positions of the X-axis and Y-axis at which the unit region of the minimum number among the number of the counted unit regions and deteimines and outputs the start positions of the judged X-axis and Y-axis addresses as the positions at which the amount of information is reduced.
  • Fig. 8 shows a flow chart of the method for detecting a position at which the amount of information is reduced in accordance with a first embodiment of the grid moving method.
  • X-axis and Y-axis grid start positions XM and YN are initialized so as to detect a position at which the amount of information is reduced in which the number of unit regions in which the image of the object having predetermined shape information exists.
  • Figs. 9A through 9F illustrate a method for adjusting the grid position so that the image of one object "a” in one image frame containing the images of a plurality of objects "a, b, c, and d" each having predetermined shape information is extracted and then exist in one image frame.
  • the grid is indicated with respect to one image frame containing the images of the plurality of the objects "a, b, c, and d" as shown in Fig. 9B and then the image of the objects is extracted.
  • One method is directed to separating and extracting so that each separate image of the objects "a, b, c, and d" can exist within one egion, and the other method is directed to extracting the image of the objects "a, b, c, and d" so that more than two images can exist within one region.
  • the image of one object "a” is extracted, and then the image of the extracted object “a” can exists within the minimum unit region 33 so as to detect a position at which the amount of information is reduced.
  • the entire grid with respect to the image of the extracted object "a” is called a first extraction grid.
  • the first extraction grid with respect to the image of the extracted object "a” is segmented into a plurality of unit regions as shown in Fig. 9C.
  • the unit region positioned at the left side and uppermost portion of the first extraction grid refers to the minimum unit region 33.
  • the image of the object "a” exists in 14 unit regions.
  • a step S14 the grid is reformed by moving the X-axis grid start position XM along the X-axis in the minimum unit region by a predetermined distance M K", which is same as a length of a pixel along the X-axis, and in a step S15, it is judged as to whether the X-axis grid start position is moved along the X-axis M-times.
  • step S15 when the X-axis grid start position XM is not moved along the X-axis M-times, the steps S12 through S15 are performed, and then the X-axis grid start position XM is moved along the X-axis by a distance "K", and the number of the unit regions in which the image of the object "a" exists is counted and then stored. The above routines are repeated.
  • step S15 the entire X-axis grid start position XM is moved M-times, the X-axis grid start position XM in which the unit region having the minimum number in a step S16 is counted is determined as an optimum X-axis grid start position XoM.
  • the counting of the number of the unit regions is completed in the step S17, in a step S18, the number of the counted unit regions are stored.
  • a step S19 the Y-axis grid start position YN is moved along the Y-axis by a predetermined distance "L" which is same as a length of a pixel along the Y-axis, and in a step S20, it is judged as to whether the Y-axis grid start position YN is moved along the Y-axis N-times.
  • step S20 when the Y-axis grid start position YN is not moved along the Y-axis N-times by a predetermined distance "L”, the steps S17 through S20 are performed.
  • YN is moved based on the optimum X-axis grid start position XoM by a predetermined distance "L”, and the number of the unit regions in which the image of the object "a” exists is counted and then stored. The above routines are repeated.
  • step S20 when the Y-axis grid start position YN is moved by a predetermined distance "L" N-times, the Y-axis grid start position XM in which the unit region having the minimum number is counted is determined as the optimum Y-axis grid start position YoN, and the optimum X-axis grid start position XoM and the optimum Y-axis grid start position YoN which are determined in the steps S16 and S21 are outputted to define the position "B" at which the amount of information is reduced.as shown in Fig.9D.
  • the position ⁇ " is an intersecting position of the optimum X-axis grid start position XoM and the optimum Y-axis grid start position YoN .
  • the first embodiment of the present invention of Fig. 8 is directed to the X-axis grid start position XM, at which the minimum number of the unit regions, in which the image of the object exists, is counted by moving the entire grid along the X-axis by a predetermined distance "K" M-times, is counted, as the optimum X-axis grid start position XoM.
  • the Y-axis grid start position YN at which the unit regions having the minimum number in which the image of the object exists is counted by moving the entire grid along the Y-axis by a predetermined distance "L" N-times based on the determined optimum X-axis grid start position XoM is determined as the optimum Y-axis grid start position YoN.
  • the determined X-axis grid start position XoM and the optimum Y-axis grid start position YoN are outputted to define the position at which the amount of information is reduced.
  • the first embodiment of the present invention is directed to moving the entire grid along the X-axis M-times and moving the entire grid along the Y-axis N-times. Namely, the X-axis and Y-axis grid start positions XM and YN are moved M + N times, and then a grid start position "B"defined by the optimum X-axis and Y-axis grid start positions XoM and YoN is detected and outputted.
  • Fig. 10 shows a method for detecting a position at which the amount of information is reduced in the compaction method according to a second embodiment of the present invention.
  • a step S34 the X-axis grid start position XM is moved along the
  • the X-axis grid start position XM is moved along the X-axis more than M-times, and then it is judged as to whether it is moved more than the size of a unit region.
  • the steps S32 through S35 are performed.
  • the X-axis grid start position XM is moved along the X-axis by a predetermined distance "K", and the number of the unit regions in which the image of the object exists is counted and stored. The above routines are repeated.
  • the Y-axis grid start position YN is moved along the Y-axis by a predetermined distance
  • a step S37 it is judged as to whether the Y-axis grid start position YN is moved along the Y-axis by a predetermined distance "L" more than N-times.
  • the steps S32 through S37 are performed.
  • the Y-axis grid is moved along the Y-axis by a predetermined distance "L”
  • the X-axis grid is moved along the X-axis within the range of the unit region by a predetermined distance "K”
  • the number of the unit regions in which the image of the object exists is counted and stored. The above routines are repeated.
  • the X-axis grid start position XM and the Y-axis grid start position YN at which the unit region having the minimum number in which the image of the object exists is counted are determined as an optimum X-axis grid start position XoM and an optimum Y-axis grid start position YoN, and then the determined optimum X-axis and Y-axis grid start positions XoM and YoN are outputted to define the position at which the amount of information is reduced.
  • the second embodiment of the present invention of Fig. 10 is directed to moving the entire grid along the X-axis by a predetermined distance "K" M-times and to moving the entire grid along the Y-axis by a predetermined distance "L” M-times, and further moving the entire grid along the X-axis by a predetermined distance "K” M-times and to moving the entire grid along the X-axis by a predetermined distance "L” M-times so as to count the number of the unit regions in which the image of the object exists.
  • the X-axis and Y-axis grid start positions XM and YN at which the unit region having the minimum number is counted are determined as the optimum X-axis and Y-axis grid positions to define a position at which the amount of information is reduced and then are outputted.
  • the second embodiment of the present invention of Fig. 10 is directed to moving the X-axis and Y-axis grid start position XM and YN by a predetermined distance "K" and "L” M x N times and outputting the X-axis and Y-axis grid start positions XM and YN, at which the minimum number of unit regions is counted.
  • the embodiments of the present invention of Figs. 8 and 10 are directed to moving the grid start position along the X-axis by a predetermined distance "K", to moving the grid start position "A" along the Y-axis by a predetermined distance "L” and to detecting the optimum X-axis and Y-axis grid start positions XoM and YoN.
  • the embodiment of the present invention of Figs. 8 and 10 are further directed to gradually moving to the Y-axis by a predetermined distance "L" and to the X-axis by a predetermined distance "K” and detecting the optimum X-axis and Y-axis grid start positions XoM and YoN and outputting the position at which the amount of information is reduced.
  • the first and second embodedments of the grid moving methods as shown in Figs. 8 and 10, a way of determining the grid start position of X-axis first followed by determining the grid start position of Y-axis is described for illustrative purpose.
  • the grid moving methods can also be processed by determining the grid start position of Y-axis first followed by determing the grid start position of X-axis.
  • Fig. 11 shows a method for detecting a position at which the amount of information is reduced in the grid moving method according to a third embodiment of the present invention.
  • a step S41 the X-axis and Y-axis grid start positions XM and YN are initialized to 0 so as to detect the optimum X-axis and Y-axis grid start positions XoM and YoN at which the number of the unit regions in which the image of the object exists becomes minimum.
  • step S42 the number of the unit regions in which the image of the object exists is counted from the initialized X-axis and Y-axis grid start positions XM and YN, and in a step S43, the number of the counted unit regions is stored.
  • a step S44 the X-axis grid start position XM is moved along the X-axis M-times, and it is judged as to whether the Y- axis grid start position YN is moved along the Y-axis N-times.
  • step S45 when the X-axis grid start position XM is not moved along the X-axis more than M-times or when the Y-axis grid start position YN is not moved along the Y-axis more than N-times, in step S45, the X-axis and Y-axis grid start positions XM and YN are moved within the unit region in a zig-zag manner by predetermined distances "K" and "L”, and then the steps S42 through S45 are performed. Thereafter, the number of the unit regions in which the image of the object exists is counted and stored. The routines of moving the X-axis and Y-axis grid start positions XM and YN are repeated.
  • Figs. 12A and 12B there are two methods of moving the X-axis and Y-axis grid start positions XM and YN in a zig-zag manner by predetermined distances "K" and "L". For example, as shown in Figs. 12A and 12B, there is shown a first method for moving to the direction indicated by the arrow and a second method for moving to the direction indicated by the arrow.
  • step S44 when the X-axis grid start position XM is moved along the X-axis M-times, and the Y-axis grid start position YN is moved along the Y-axis N-times, in a step S46, the X-axis grid start position XM and the Y-axis grid start position YN at which the unit regions of the minimum number are counted are determined as the optimum X-axis grid start position XoM and the optimum Y-axis grid start position YoN, and then the determined optimum X-axis grid start position XoM and the determined optimum Y-axis grid start position YoN are outputted to define the positoin "B M .
  • another embodiment of the present invention of Fig. 11 is directed to moving the entire grid within the range of the X-axis and Y-axis of the unit region in a zig-zag manner and to counting the number of the unit regions in which the image of the object exists. Therefore, the third embodiment of the present invention of Fig. 11 is directed to moving the X-axis and Y-axis grid start positions XM and YN by predetermined distances "K" and "L” M x N times and determining the X-axis and Y-axis grid start positions XM and YN, at which a minimum number of unit regions is counted, as the position B",as shown in Fig.9D, at which the amount of information is reduced.
  • Fig. 9D shows an example for reforming the grid to the position at which the amount of information is reduced which is determined in accordance with the embodiments of the present invention of Figs. 8, 10, and 11.
  • the 45° rotated square, the hexagonal- shaped form or the like may be used.
  • more than two different shapes which are capable of dividing the image frame into a predetermined shape at a regular interval may be combined so as to form the unit regions.
  • Fig. 13 shows a flow chart of a method for detecting the position at which the amount of informationis reduced in a grid moving method in accordance with a fourth embodiment of the present invention.
  • a step S51 the X-axis and Y-axis grid start positions XM and YN are initialized as 0 so as to detect the position at which the number of the unit regions in which the image of the object exists, becomes minimum.
  • a step S54 the Y-axis grid start position YN is moved along the Y-axis by a predetermined distance "L”, and in a step S55, it is judged as to whether the Y-axis grid start position YN is moved along the Y-axis more than N-times.
  • the steps S52 through S55 are performed.
  • the routines that the Y-axis grid start position YN is moved along the Y-axis by a predetermined distance "L", and the number of the unit regions in which the image of the object exists is counted and stored are repeated.
  • step S56 the Y-axis grid start position YN at which a minimum number of unit regions exists is determined as the optimum Y-axis grid start position YoN.
  • step S57 the grid is reformed based on the determined optimum Y-axis grid start position YoN, and in step S58, the unit region of the current X-axis row in which the image of the object exists is counted.
  • step S59 the number of the counted unit regions is stored.
  • step S60 the grid start position XM of the X(1) row is moved along the X-axis by a predetermined distance "K”, and in step S61 it is judged as to whether the X-axis grid start position XM of the X(1) row is moved along the X-axis by a predetermined distance "K" more than M-times.
  • the steps S58 through S61 are performed, and the routines that the X-axis grid start position XM of the X(1) row is moved along the X-axis by a predetermined distance "K", and the unit region in which the image of the object among the unit regions of the X(1) row exists is counted and stored are repeated.
  • step S61 when the grid start position XM of the X(1) row is moved by a predetermined distance M-times, in a step S62, the grid start position XM of the X(1) row at which the minimum number of unit regions among the currently counted unit regions is counted is determined as the optimum X(1) row grid start position X1M.
  • step S63 it is judged as to whether the row refers to the last row of the X-axis, and when the row is not referred to as the last row of the X-axis, in a step S64, the row is moved along the next row, and then the steps S58 through S64 are performed.
  • the above routines are repeatedly performed, and the row of the X-axis is sequentially moved along X(1), X(2), X(3), X(4), and X(5), and the grid start position XM of the X(1), X(2), X(3), X(4), and X(5) rows at which the minimum number of unit regions in which the image of the object exists is counted are determined as the grid start positions (X1M, X2M, ...) of the optimum X(1), X(2), X(3), X(4), and X(5) rows.
  • the optimum Y-axis grid start position YoN when the row refers to the last row of the X-axis, the optimum Y-axis grid start position YoN, the optimum grid start positions (X1M, X2M, ...) of the X(1), X(2), X(3), X(4), and X(5) are outputted as positions at which the amount of the information is reduced.
  • Fig.14 shows the flow chart of a method for detecting the positions at which the amount of information is reduced in the grid moving method in accordance with a fifth embodiment of the present invention.
  • step S71 the X-axis and Y-axis grid start positions XM and YN are initialized as 0 so as to detect the positions at which the number of the unit regions in which the image of the object exists becomes minimum.
  • step S72 the unit region in which the image of the object exists among the unit regions of the current X-axis row is counted.
  • the unit region in which the image of the object exists among the unit regions of the X(1) row is detected and counted.
  • step S73 the number of counted unit regions is stored.
  • step S74 the X-axis grid start position XM of the X(1) row is moved along the X-axis by a predetermined distance "K", and in step S75, it is judged as to whether the X-axis grid start position XM of the X(1) row is moved by a predetermined distance "K" more than M-times.
  • step S75 when the grid start position XM of the X(1) is not moved by a predetermined distance "K" more than M-times, the steps S72 through S75 are performed.
  • step S75 when the X-axis grid start position XM of the X(1) row is moved along the X-axis by a predetermined distance "K" more than M-times, the grid start position XM of the X(1) row at which the minimum number of unit regions among the currently counted unit regions of the X(1) is counted is determined as the optimum X(1) row grid start position X1M.
  • step S77 it is judged as to whether the row refers to the last row of the A-axis, and when the row is not the last row of the X-axis, in step S78, the row is sequentially moved along the next row of the X-axis, namely, to the X(2), X(3), X(4), and X(5) rows, and the steps S72 through S78 are performed, and the routines that the grid start position XM of the X(2), X(3), X(4), and X(5) at which the minimum number of unit regions in which the image of the object exists is counted is determined as the grid start positions (X1M, X2M, ...) of the optimum X(2), X(3), X(4), and X(5) rows are repeatedly performed.
  • step S77 when the row refers to the last row of the X-axis, in step S79, it is judged as to whether the Y-axis start position YN is moved along the Y-axis by a predetermined distance "L" more than N-times.
  • step S79 when the Y-axis grid start position YN is not moved along the Y-axis by a predetermined distance "L" more than N-times, in step S80, the Y-axis grid start position YN is moved along the Y-axis by a predetermined distance "L”, and the steps S72 through S80 are repeatedly performed.
  • the number of the unit regions in which the image of the object exists is counted by moving the Y-axis grid start position YN along the Y-axis by a predetermined distance L" and by sequentially moving the grid start position XM of the X(1), X(2), X(3), X(4), and X(5) by a predetermined distance "K" from the position to which the Y-axis grid start position YN is moved.
  • the grid start position XM of the X(1), X(2), X(3), X(4), and X(5) rows at which the minimum number of unit regions is counted is sequentially determined as the grid start positions (X1M, X2M, ...) of the optimum X(1), X(2), X(3), X(4), and X(5) rows.
  • step S81 when the Y-axis grid start position YN is moved by a predetermined distance "L" more than N-times, in step S81, the numbers of the unit regions which are counted at the grid start positions (X1M, X2M, ...) of the optimum X(1), X(2), X(3), X(4), and X(5) rows which are determined at the position to which the Y-axis grid start position YN is moved are all summed.
  • step S82 as a result of the sum, the position at which the minimum number of unit regions is counted is judged as the Y-axis grid start position YN, and the judged Y-axis grid start position YN is determined as the optimum Y-axis grid start position YoN.
  • the grid start positions (X1M, X2M, ...) of the X(1), X(2), X(3), X(4) and X(5) rows at which the minimum number of unit regions of the minimum number is counted is judged as the optimum Y-axis grid start position YoN are determined as the grid start positions(X1M, X2M, ...) of the optimum X(1), X(2), X(3), X(4) and X(5) rows.
  • the determined optimum Y-axis grid start position YoN and the grid start positions (X1M.X2M, ...) of the X(1), X(2), X(3), X(4), and X(5) are outputted as the position at which the amount of information is reduced.
  • Fig. 9E The result of the reformation of the grid in accordance with the position at which the amount of information is reduced which is obtained by the embodiments of Figs. 13 and 14 are shown in Fig. 9E.
  • the present invention is directed to detecting the position at which the amount of informationis reduced in which the image of the object exists in the minimum number of unit regions by changing the X-axis grid start position XM and the Y-axis grid start position YN as shown in Figs. 13 and 14.
  • the grid moving methods can also be processed by determining the grid start position of each column after determining the optimum X-axis grid start position.
  • the unit region may be formed in various shapes.
  • the unit regions may be moved by separating the row of the X-axis or the row of the Y-axis.
  • the unit region may be formed in a 45° rotated square shape and then the unit region is segmented by a slant grid.
  • the unit region may be also formed of movable slant grid.
  • a unit region may be formed as a triangle using two slant grids 15 and 17 opposed to the X-axis grid 11.
  • a unit region may be formed as a triangle using two slant grids 15 and 17 opposed to the Y-axis grid 13.
  • the unit region may be formed as a 45° rotated square using two opposed slant grids 15 and 17.
  • a method of extracting the image of the object "a" and positioning it into the minimum number of unit regions has been described.
  • Y-axis grid start positions XM and YN have the reference of the number of the pixels existing in the range of the unit region.
  • the X-axis and Y-axis grid start positions XM and YN may be moved by the length of the unit pixel within the range of the
  • the moving distances K and L" of the X-axis and Y-axis grid start positionin XM and YN consideration of the information with respect to the chrominance signal and the luminance signal are preferably referred to as the length of two pixels.
  • the unit region is divided into sub-regions having the size of (M/2) X (N/2), and the grid start positions XM and YN are moved by predetermined distances "K and L" within the range of the X-axis and Y-axis of the divided sub-regions, and the X-axis and Y-axis grid start positions at which the image of the object exists in the minimum number of unit regions is detected and outputted as the position at which the amount of information is reduced.
  • the sub-region has 8 X 8 pixels.
  • the grid start positions XM and YN are moved by predetermined lengths "K and L" within the number of pixels of the sub-region which is divided into 8 X 8 pixels.
  • the X-axis and Y-axis grid start positions at which the minimum number of unit regions in which the image of the object exists is detected, and then the X-axis and Y-axis grid start positions are outputted as the positions at which the amount of information is reduced .
  • a plurality of positions at which the amount of information is reduced for counting the minimum number of unit regions may be generated.
  • the value of the motion vector becomes smaller, and the information amount is decreased.
  • Figs. 15 and 16 were obtained by adjusting the positions of the X-axis and y-axis grids with respect to the image of the object and coding using a computer simulation.
  • the image used for the object was that of a young woman and an old woman, and the number of the image frames was 50, respectively.
  • a fixed block grid(FBG) shape adaptive discrete cosine transform refers to a method of adjusting the position of the X-axis grid and Y-axis grid and coding in accordance with the first.second and third embodiments of Figs. 8, 10, and 11.
  • a method of adjusting and coding the position of the X-axis grid and Y-axis grid in accordance with the X-axis grid and Y-axis grid refers to a variable block grid (VBG) shape adaptive discrete cosine transform (SADCT) 1-X method and a variable block grid (VBG) shape adaptive discrete cosine transform (SADCT) 1-Y method.
  • VBG variable block grid
  • SADCT shape adaptive discrete cosine transform
  • a method of adjusting and coding the position of the X-axis grid and Y-axis grid in accordance with the X-axis grid and Y-axis grids refers to a variable block grid (VBG) shape adaptive discrete cosine transform (SADCT) 2-X method and a variable block grid (VBG) shape adaptive discrete cosine transform (SADCT) 2-Y method.
  • VBG variable block grid
  • SADCT shape adaptive discrete cosine transform
  • a method of separating and extracting the image of the object and then coding in accordance with the X-axis grid is referred to a variable block grid (VBG) shape adaptive discrete cosine transform (SADCT) 2-X method (by the object).
  • an image of the object between N-1 and N frames of the original image is extracted, and the result between the conventional compaction method which is directed to performing the shape adaptive discrete cosine transform (SADCT) without varying the position of the X-axis grid and Y-axis grid from the extracted image and the compaction method of the present invention was analyzed.
  • SADCT shape adaptive discrete cosine transform
  • the comparison method was conducted by judging how much transmission bit rate occurs as the occurrence bits per pixel(BPP) per frame and the number of regions per frame when the pixel of the object is identical with the peak signal to noise ratio(PSNR) which is the objective picture quality evaluation reference.
  • Fig. 15 shows the average result which is obtained by an experiment of 50 frames with respect to the image of the young woman.
  • the average peak signal to noise ratio(PSNR) value was 36.46db
  • the number of the average occurrence bits was 751
  • the number of the average BPP was 1.21
  • the number of the average unit regions was 20.71.
  • the FBG-SADCT was performed after reforming the grid using the X-axis and Y-axis which is moved so that the image of the object can exist in the minimum number of unit regions.
  • the average PSNR was 36.37db, and the number of the average occurrence bits was 719, and the average BPP was 1.16, and the number of the average unit regions was 18.65.
  • the FBG- SADCT of the present invention had the average number of the occurrence bits reduced by 32 bits, the BPP was reduced by 0.05, and the number of the average unit regions was reduced by 2.06.
  • the average PSNR value was 36.3db, and the number of the average occurrence bits was 694, and the average BPP was 1.12, and the number of the average regions was 16.82.
  • the present invention it is possible to reduce the average occurrence bit rate by 57 bits performing the VBG-SADCT by the image of the object after adjusting the position of the grid in accordance with the position in which the image of the object exists compared to the conventional method which is directed to performing the SADCT without moving the position of the grid in accordance with the position at which the image of the object exists.
  • the average BPP is reduced by 0.09, and the number of the average unit regions is reduced by 3.89.
  • Fig. 16 shows the result which is obtained by performing the SADCT without varying the grid in the conventional method with respect to the image of the object among the image of the old woman and the result which is obtained by performing the FBG-SADCT and the VBG-SADCT after varying the position of the grid in the compaction method according to the present invention.
  • Fig. 17 shows a block diagram showing the compaction/motion estimation apparatus according to a first embodiment of the present invention, which is directed to detecting the grid moving position and the position at which the amount of information is reduced with respect to the image of the object having predetermined shape information and estimating the compaction and motion after reforming the grid in accordance with the position at which the amount of informationis reduced.
  • reference numeral 41 denotes an image signal input unit for inputting the image of the object having predetermined shape information.
  • Reference numeral 43 denotes a grid moving unit for moving the position of the X- axis grid and the Y-axis grid in accordance with the position in which the image of the object exists and for detecting a position at which the image of the object exists in the minimum number of unit regions.
  • the grid moving unit 43 having the same construction as the grid moving apparatus as shown in Fig. 4 is directed to sequentially storing the image of the object which is inputted by the image signal unit 41 , moving the grid start position with respect to the stored image of the object at a predetermined distance within the range of the X-axis and Y-axis, and separating the image of the object into a plurality of unit regions in accordance with the moved grid start position.
  • the grid moving unit 43 is directed to judging and counting the unit regions in which the image of the object exists among the unit regions and outputting the X-axis and Y-axis grid start positions at which the minimum number of unit regions among the counted values is counted as the positions at which the amount of information is reduced.
  • reference numeral 45 denotes a compaction unit for reforming the X-axis grid and the Y-axis grid in accordance with the positions at which the amount of information is reduced outputted from the grid moving unit 43 and for coding the image of the object.
  • the compaction unit 45 is directed to reforming the grid in accordance with the positions at which the amount of information is reduced, namely, which is referred to as the X-axis grid start position and Y-axis grid start position at which the image of the object exists in the minimum number of unit regions.
  • the positions at which the amount of information is reduced is detected by the grid moving apparatus and the grid moving method.
  • the compaction is performed using SADCT, DCT, vector quantumization or the like.
  • reference numeral 47 denotes a motion estimation unit for estimating the motion of the image of the object.
  • the motion estimation unit 27 is directed to reforming the grid in accordance with the positions at which the amount of information is reduced, which is detected by the grid moving apparatus and the grid moving method, segmenting the unit regions in which the image of the object exists in the reformed grid, and generating the motion information by estimating the varied amount of the segmented unit regions.
  • the present invention is directed to estimating the motion by reforming the grid from the above- mentioned object as shown in Fig. 18C in accordance with the positions at which the amount of information is reduced which are detected by the grid moving apparatus and the grid moving method.
  • Fig. 19 shows a compaction/motion estimation apparatus according to a second embodiment of the present invention which is directed to separating the image of the object having predetermined shape information into the motion compensable object and the motion compensable failed object, coding the separated motion compensable failed object, and estimating the motion of the motion compensable object.
  • reference numeral 51 denotes an image signal input unit for inputting an image signal having predetermined shape information to be coded.
  • Reference numeral 53 denotes an image separation unit for separating the image signal outputted from the image signal input unit 51 into the image of the background image having no motion and the image of the moving object having motion.
  • the changed region of the image of the moving object is judged using the information varied between the previously inputted image and the currently inputted image.
  • Fig. 20 shows the time-based variation of the inputted image frame. As shown therein, it appeared that a plurality of intermediate image frames Bi, ..., Bj exist between the frame P of the previously inputted image and the frame P of the currently inputted image.
  • the image of the moving object estimated by the image separation unit 53 is separated into moving information of the motion compensable object, shape information of the motion compensable object, signal information of the motion compensable failed object, and shape information of the motion compensable failed object.
  • reference numeral 55 denotes a first grid moving unit for detecting the positions at which the amount of information is reduced using shape information of the motion compensable failed objectseparated by the image separation unit 53.
  • the first grid moving unit 55 has the same construction as the grid moving unit 43 and as the grid moving apparatus of Fig. 4.
  • the first grid moving unit 55 is directed to adjusting the grid position in accordance with the position at which the image of the motion compensable failed object separated by the image separation unit 53 and outputting the position in which the image of the motion compensable failed object exists in the minimum number of unit regions.
  • reference numeral 57 denotes a compaction unit for coding a signal of the position at which the amount of information is reduced outputted from the first grid moving unit 55 and the image of the motion compensable failed object in accordance with the position at which the amount of informationis reduced detected the grid moving method.
  • the compaction unit 57 is directed to reforming the grid using the position at which the amount of information is reduced as a reference and coding the image of the motion compensable failed object of the unit region in which the image of the motion compensable failed object exists in the unit region of the reformed grid using the methods of SADCT, DCT, or vector quantumization.
  • reference numeral 59 denotes a second grid moving unit having the same construction as the grid moving apparatus of Fig. 1 , the grid moving unit 43, and the first grid moving unit 55.
  • the first grid moving unit 55 is directed to adjusting the position of the grid in accordance with the position at which the image of the motion compensable object separated by the image separation unit 53 and outputting the position in which the image of the motion compensable object exists in the minimum number of unit regions.
  • Reference numeral 61 denotes a motion estimation unit which is directed to reforming the grid in accordance with the signal of the position at which the amount of information is reduced outputted from the second grid moving unit 59 and the position at which the amount of informationis reduced detected by the grid moving method, segmenting the image of the motion compensable object into the unit regions from the reformed grid, and generating the motion information using the variation amount of the segmented unit regions.
  • the grid moving method for minimizing image information of an object and an apparatus using the grid moving method and the compaction/motion estimation method using the grid moving method and an apparatus thereof according to the present invention are directed to forming the minimum number of unit regions having the image of the object by adjusting the position of the grid in accordance with the image position of the object, coding the image of the object with respect to the minimum number of unit regions , and generating the motion information, thus achieving a higher compaction rate, whereby the amount of the data to be stored and transmitted can be significantly reduced.

Abstract

An improved grid moving method of an object image and an apparatus using the same which are capable of reducing the amount of information with respect to the image of an object by moving the grid in accordance with a position in which an image of the object having shape information exists, which include the steps of: a moving step for forming a grid over an image of an object having shape information, segmenting the image into a plurality of unit regions, and moving the formed grid; a judging step for judging an amount of the information at each position to which the grid is moved in the moving step; a detecting step for detecting a position at which the amount of the information is reduced; a compaction step for reforming the grid in accordance with the position detected in the detecting step and for coding the image of the object existing in unit regions of the reformed grid; and a motion estimation step for reforming the grid in accordance with the position detected in the detecting step and for estimating the motion of the image of the object in the unit regions in which the image of the object exists among the unit regions segmented by the reformed grid.

Description

GRID MOVING METHOD OF OBJECT IMAGE AND APPARATUS USING
THE SAME AND COMPACTION/MOTION ESTIMATION METHOD USING
THE SAME AND APPARATUS THEREOF
TECHNICAL FIELD
The present invention relates to a grid moving method for inimizing image information of an object image and an apparatus therefore and a compaction/motion estimation method using the grid moving method and an apparatus therefor, and particularly to an improved grid moving method for an object image and an apparatus therefor and a compaction/motion estimation method using the grid moving method and an apparatus therefor which are capable of forming a grid with respect to an image of a predetermined object having shape information and.in a region of an image, dividing the region of the image into a plurality of unit regions, moving the formed grid, and detecting a position at which the amount of information is reduced when performing a compaction or estimating a motion of an object. In addition, the present invention is basically directed to reforming the grid to a position at which the amount of information is reduced by moving the grid, and is directed to separating and coding each unit region in which an image of the object exist from the reformed grid and using the grid movement of the image of an object of which the motion is estimated.
BACKGROUND ART
Conventionally, since an image of an object having a predetermined shape contains a great amount image data, when storing the image data in a recording/writing medium, a large space is necessary for storing the data. In addition, transmitting the data takes too much time, so it is difficult to transmit the data in real time.
Therefore, the image of the object is coded, and the motion of the image is estimated and then the amount of the information in the image is reduced for storing the information in a predetermined recording/writing medium. Thereafter, the information is transmitted to a predetermined destination in real time.
When coding an image of an object, a vector quantum ization method or a discrete cosine transform (DCT) method is used.
Recently, a shape adaptive discrete cosine transform (SADCT) method has been effectively used in industry. This method is very effective for object- based compaction.
The above-mentioned shape adaptive discrete cosine transform method is directed to forming a grid with respect to an image frame, dividing an image of an object into a plurality of unit regions each having predetermined size and shape information, separating a unit region from the plurality of the unit regions in which an image of the object exists and then coding the unit region. In addition, when the unit region contains an image to be coded, the effectiveness between a two-dimensional region DCT and a compaction become identical in the shape adaptive discrete cosine transform. When the unit region does not contain the image to be coded, the pixel, in which an image of an object exists, is processed with respect to the X-axis in a one-dimensional discrete cosine transform method, and a result of the above X-axis-based process is processed with respect to the Y-axis in a one-dimensional discrete cosine transform method. Thereafter, the final result value is obtained.
The shape adaptive discrete cosine transform method is further directed to reducing the number of unit regions in which an image of an object exists and performing the compaction after substantially filling the image of the object in the unit region, thus enhancing the compaction of a transform constant.
Therefore, when performing the shape adaptive discrete cosine transform process, the image of the object to be coded should preferably be substantially filled in each unit region, and then the number of unit regions in which an image of the object exists is effectively reduced.
The above- described shape adaptive discrete cosine transform process will now be described in more detail with reference to Figs. 1A through 3F. Figs. 1A and 1B show grid patterns formed in one frame.
As shown therein, one frame is divided into a plurality of rows and columns which are consisted of a plurality of unit regions 21 having the same size and shape in cooperation with a P X Q number of X-axis grid and Y-axis grid 11 and 13 spaced apart from one another at a regular distance.
A unit region 21 may be formed in various shapes.
For example, the unit region 21 is formed in a regular square or a rectangular form by the X-axis and Y-axis grid 11 and 13. In addition, as shown in Fig. 2A, A unit region 21 may be formed as a horizontally lying triangle or a horizontally upside down triangle, and neighboring triangles form rectangular shapes bounded by the slant grids 15 and 17. As shown in Fig. 2B, a unit region 21 is formed by vertically lying triangles and neighboring triangles form rectangular shapes bounded by the slant grids 15 and 17.
In addition, as shown in Fig. 2C, the unit region 21 is formed as a 45° rotated square by the slant grids 15 and 17, and as shown in Figs. 2D and 2E, the unit region 21 is foimed in a hexagonal shape by the slant grids 15 and 17. As shown in Fig. 2F, the unit region 21 is formed in an octagonal form having a 45° rotated small square between the neighboring octagons. In this example, two different shaped unit regions 21 are concurrently used. Any shape which spatially and evenly divides the image frame may be used for the unit region 21.
A square- or rectangular-shaped unit region 21 which is defined by an X-axis grid 11 and Y-axis grid 13 will now be explained.
As shown in Fig. 1B, the unit region 21 is formed of an M X N number of unit pixels 23 in the X-axis and Y-axis directions. For example, one unit region 21 is formed of an 8 X 8 number of unit pixels 23 or is formed of a 16 X 16 number of unit pixels 23.
In addition, a unit region 21 is defined as an M X N number of blocks in accordance with the number of unit pixels 23. As shown in Fig. 1B, the unit region 21 refers to an 8 X 8 number of blocks corresponding to unit pixels.
Fig. 3A shows an image (shown as the hatched portion) having predetermined shape information in a unit region 21 formed of an 8 X 8 number of unit pixels 23. For the shape adaptive discrete cosine transform with respect to the image of an object, as shown in Fig. 3B, the image of the object is filled from the upper side margin portion of the unit region 21, and then the one-dimensional cosine transform is performed with respect to the Y-axis which is shown in the vertical direction.
The one-dimensional discrete cosine transform is performed as shown in Fig. 3D. When the one-dimensional discrete cosine transform is completed with respect to the Y-axis, the image of the object is filled from the left side margin portion of the unit region 21, as shown in Fig. 3E,and then the one-dimensional discrete cosine transform is performed with respect to the X-axis which is shown in the horizontal direction. When the one-dimensional discrete cosine transform is completed with respect to the X-axis, as shown in Fig. 3F, the shape adaptive discrete cosine transform with to the Y-axis and X-axis is completed.
Thereafter, a zig-zag scan is performed with respect to the final shape, as shown in Fig. 3F, which is obtained by the above-mentioned shape adaptive discrete cosine transform. For example, the zig-zag scan is performed diagonally from the leftmost side and the uppermost side to the rightmost side and the lowermost side.
However, the conventional shape adaptive discrete cosine transform is directed to performing the shape adaptive discrete cosine transform in accordance with the position in which the image of an object exists without moving the position of the grid.
Therefore, the bit rate per frame is high, and since the number of the unit regions in which the image of the object exists is numerous, there is a restriction on the ability to reduce an amount of compaction information which is obtained by coding the image of the object and the amount of motion information which is obtained by estimating the motion of the object.
In addition, when coding an object in the conventional discrete cosine transform method or the vector quantum ization, since the compaction is performed without moving the position of the grid in accordance with the position in which the image of the object exists, the bit rate per frame is high as in the shape adaptive discrete cosine transform, and since the number of unit regions in which the image of the object exists is numerous, there is a restriction on the ability to reduce the amount of compaction information and the amount of motion information.
Meanwhile, when coding the image of a moving object among the images of an object having predetermined shape information, an object-based moving image coding method is generally used in the industry.
The above-mentioned object-based moving image coding method is directed to segmenting the image of the object in a background in which there is not a moving image and a changed region which is defined by the moving image of the object.
In addition, the moving object of the changed region is segmented into a motion compensable object and a motion compensable failed object through motion estimation.
Here, the motion compensable object refers to the moving object having a predetermined theory such as a horizontal movement, a rotational movement, a lineal movement, and the like in a state that the object in a three-dimensional space is converted into a two-dimensional image of the object. In addition, the motion compensable failed object refers to an object which is not adaptable with respect to the above-mentioned theory.
When transmitting and storing the image of the object, the motion compensable object process is directed to detecting motion information of the image of the object.
In addition, the image of the motion compensable failed object and the image of the exposed object are most effectively coded so as to reduce the amount of information, which is then transmitted and stored. Since the amount of information with respect to the image of the motion compensable failed object is about 60-70% of the total amount of the information to be transmitted, many studies have been conducted, in the industry, so as to reduce the amount of information transmitted.
The motion estimation of the motion compensable object is directed to segmenting and estimating the moving portion of the moving image from a picture of the previous frame so as to minimize the amount of motion information.
However, since the variables with respect to the moving object are various, it is difficult to effectively extract, transmit, and store motion information in response to the immediate movement of the object.
Therefore, in the industry, it is urgently needed to transmit and store picture information of a high resolution having a small amount of information with respect to the motion compensable object in the motion estimation method.
DISCLOSURE OF THE INVENTION Accordingly, it is an object of the present invention to provide a grid moving method for minimizing image information of an object image and an apparatus using the grid moving method and a compaction/motion estimation method using the grid moving method and an apparatus therefor which overcome the problems encountered in the conventional method of an object image and an apparatus therefor and a conventional δ compaction/motion estimation method and an apparatus therefor.
It is another object of the present invention to provide a grid moving method for an object image and an apparatus using the grid moving method which are capable of reducing amount of information with respect to the image of an object by moving the grid in accordance with a position in which an image of the object having predetermined shape information exists.
It is another object of the present invention to provide a grid moving method for an object image and an apparatus using the grid moving method which are capable of reducing the number of unit regions in which an image of the object exists by moving the grid so as to reduce the amount of the information with respect to the image of the object.
It is another object of the present invention to provide an image signal coding apparatus and a compaction method using a grid moving method which are capable of reducing the amount of compaction information by compacting the unit regions in which the image of the object exists among the unit regions segmented by the grid moved in accordance with the position in which the image of the object exists. It is another object of the present invention to provide an image signal coding apparatus and a compaction method using a grid moving method which are capable of reducing the amount of information by compacting the unit region in which the image of a motion compensable failed object having predetermined shape information exists among the unit regions which is segmented by the grid moved in accordance with the position in which the image of the motion compensable failed object exists.
It is another object of the present invention to provide a motion estimation apparatus and a method of the motion estimation apparatus using a grid moving method which is capable of estimating the motion information as a unit region in which an image of the motion compensable object exists among the unit region which is segmented by the grid which is moved in accordance with the position in which an image of the motion compensable object exists.
To achieve the above objects, the present invention is basically directed to segmenting an image of an object having predetermined shape information with a grid, detecting a position in which the amount of this information can be reduced by moving the position of the grid along the X-axis or Y-axis directions, and moving the grid to the position in which the amount of information can be reduced.
To achieve the above objects, the present invention is directed to moving the position of the grid so that the image of the object can be positioned in the minimum number of unit regions so as to reduce the amount of information with respect to the image of the object.
To achieve the above objects, the present invention is directed to detecting the number of unit regions in which the image of an object exists and moving the position of the grid so that the image of the object exists in the minimum number of unit regions and coding the image of the object existing in the detected unit regions so as to minimize the amount of information.
To achieve the above objects, the present invention is directed to judging the unit regions in which the image of the object exists in a state that the position of the grid is moved so that the image of the object can exists in the minimum number of unit regions, and estimating the motion of the object using the judged unit region and detecting the position of the grid using the information of the object.
To achieve the objects above, there is provided a grid moving method of an object image, including the steps of: a segmenting step for forming a grid over an image of an object having predetermined shape information and for segmenting the image into a plurality of unit regions; and a detecting step for detecting a position at which the amount of information is reduced by moving the grid formed in the segmenting step.
To achieve the objects above, there is provided a compaction/motion estimation method, including the steps of: a moving step for forming a grid over an image of an object having predetermined shape information, segmenting the image into a plurality of unit regions, and moving the foimed grid; a judging step for judging the amount of information at each posion to which the grid is moved in the moving step; a detecting step for detecting a position at which the amount of information is reduced; a compaction step for reforming the grid in accordance with the position detected in the detecting step and for coding the image of the object existing in each unit region of the reformed grid; and a motion estimation step for reforming the grid in accordance with the position detected in the detecting step and for estimating the motion of the image of the object in the unit regions in which the image of the object exists among the unit regions segmented by the reformed grid.
To achieve the objects above, there is provided a compaction method, including the steps of: a moving step for forming a grid over an image of an object having predetermined shape information, segmenting the image into a plurality of unit regions, and for moving the formed grid; a judging step for judging the amount of information at each position to which the grid is moved in the moving step; a detecting step for detecting a position at which the amount of information is reduced in the judging step; and a compaction step for reforming the grid in accordance with the position detected in the detecting step and coding the image of the object existing in the unit region of the reformed grid.
To achieve the objects above, there is provided a motion estimation method, including the steps of: a moving step for forming a grid over an image of an object having predetermined shape information, segmenting the image into a plurality of unit regions, and moving the formed grid; a judging step for judging the amount of information at each position to which the grid is moved in the moving step; a detecting step for detecting a position at which the amount of information is reduced in the judging step; and a motion estimation step for reforming the grid in accordance with the position detected in the detecting step and for estimating the motion of the image of the object in the unit regions in which the image of the object exist among the unit regions by the reformed grid.
To achieve the objects there is provided a compaction/motion estimation method, including the steps of: a separating step for estimating the motion of an image of an object having predetermined shape information and for separating a motion compensable failed object image and a motion compensable object image; a first moving step for forming a grid over the motion compensable failed object image separated in the separating step, segmenting the image into a plurality of unit regions, and moving the grid; a first judging step for judging the amount of information at each position to which the grid is moved in the first moving step; a first detecting step for detecting a position at which the amount of information is reduced in the first judging step; a compaction step for reforming the grid in accordance with the position detected in the first detecting step and for coding the unit regions in which the image of the motion compensable failed object exists from among the unit regions which is segmented by the reformed grid; a second moving step for forming a grid over the image of the motion compensable object separated in the separating step, segmenting the image into a plurality of unit regions, and moving the grid; a second judging step for judging an amount of information at each position to which the grid is moved in the second moving step; a second detecting step for detecting a position at which the amount of information is reduced in the second judging step; and a motion estimating step for reforming the grid in accordance with the position detected in the second detecting step and for estimating the motion of the image of the motion compensable object using the unit regions in which the image of the motion compensable object exists from among the unit regions segmented by the reformed grid.
To achieve the objects above, there is provided a compaction method using a grid moving method for an object image, including the steps of: a separating step for estimating the motion of the image of an object having shape information and for separating the image of a motion compensable failed object; a moving step for forming a grid over the image of the motion compensable failed object separated in the separating step, segmenting the image into a plurality of unit regions, and moving the grid; a judging step for judging an amount of information at each position to which the grid is moved in the moving step; a detecting step for detecting a position at which the amount of information is reduced in the judging step; and a compaction step for reforming the grid in accordance with the position detected in the etecting step and for coding the unit regions in which the image of the motion compensable failed object exists among the unit regions segmented by the reformed grid.
To achieve the objects above, there is provided a motion estimation method using a grid movement of an image of an object, including the steps of: a separating step for estimating the motion of the image of the object having shape information and for separating the image of the motion compensable object; a moving step for forming a grid over the image of the motion compensable object separated in the separating step, segmenting the image into a plurality of unit regions, and moving the grid; a judging step for judging an amount of information at the position to which the grid is moved in the moving step; a detecting step for detecting a position at which the amount of information is reduced in the judging step; and a motion estimating step for reforming a grid in accordance with the position detected in the detecting step and for estimating the motion of the image of the motion compensable object using the unit regions in which the image of the motion compensable object exists from among the unit regions segmented by the reformed grid. To achieve the objects above, there is provided a grid moving apparatus for an object image, including: an address generation controller for moving an address start position at which an address is generated by a predetermined distance within a predetermined region of a unit region; an address generator for separating the image of the object into unit regions in accordance with the address start position which the address generation controller outputs and for generating the address; a memory unit for storing the image of the object having inputted shape information and for outputting the image in accordance with an address outputted from the address generator; a region number counter for counting the number of unit regions in which the shape information of the object exists outputted from the memory unit; and minimum unit region grid selector for selecting an X-axis grid start position XM and a Y-axis grid start position YN at which the minimum number of unit regions is counted from among the number of the unit regions counted by the region number counter.
To achieve the objects above, there is provided a motion estimation apparatus using a grid moving of an object image, including: a grid moving unit for adjusting the grid in accordance with an image position of an object having shape information outputted from an image signal input unit and for reducing the number of unit regions in which the image of the object exists; and a motion estimation unit for estimating a motion of the image of object using the motion of the unit regions of which the number of unit regions is reduced.
To achieve the objects above, there is provided a compaction/motion estimation apparatus using a grid moving of an object image, including: an image separator for separating an image of a motion compensable failed object and an image of a motion compensable object in accordance with a motion from an image of an object having shape information; a first grid moving unit for adjusting the grid in accordance with an image position of the motion compensable failed object separated by the image separating unit and for reducing the number of unit regions in which the image of the motion compensable failed object exists; a compaction unit for coding the image of the object existing in the unit regions of which the number of the unit regions is reduced by the first grid moving means; a second grid moving unit for adjusting the grid in accordance with the image position of the motion compensable object separated by the image separating unit and for reducing the number of unit regions in which the image of the motion compensable object exists; and a motion estimation unit for estimating the motion information of the motion compensable object using the unit regions in which the image of the motion compensable object exists of which the number of the unit regions is reduced.
To achieve the objects above, there is provided a motion estimation apparatus using a grid moving of an object image, including: an image separating unit for separating an image of a motion compensable object in accordance with a motion of an object having shape information; a grid moving unit for adjusting a grid in accordance with an image position of the motion compensable object separated by the image separating unit and for reducing the number of unit regions in which the image of the motion compensable object exists; and a motion estimation unit for estimating motion information of the motion compensable object using the unit region in which the image of the motion compensable object exists of which the number of the unit regions is reduced by the grid moving unit.
BRIEF DESCRIPTION OF THE DRAWINGS The present invention will become more fully understood from the detailed description given hereinbelow and the accompanying drawings which are given by way of illustration only, and thus are not limitative of the present invention, and wherein:
Figs. 1A and 1B are views showing a conventional grid pattern formed in one image frame, of which: Fig. 1A is a view showing a conventional grid pattern formed in one image frame; and
Fig. 1B is a view showing a conventional unit region which is indicated as 8 X 8 pixels; Figs. 2A through 2F are views showing various shapes of a conventional unit region;
Figs. 3A through 3F are views showing a conventional shape adaptive discrete cosine transform process;
Fig. 4 a block diagram showing a grid moving apparatus according to the present invention;
Fig. 5 is a block diagram showing an address generator of Fig. 4 according to the present invention;
Fig. 6 is a view showing an order that an address generator outputs an image of an object stored in a memory of Fig. 4 by separating in accordance with an X-axis address and a Y-axis address according to the present invention;
Fig. 7 is a block diagram showing a region number counter of Fig. 4 according to the present invention;
Fig. 8 is a flow chart showing a method for detecting a position at which the amount of information is reduced in a grid moving method according to a first embodiment of the present invention;
Figs. 9A through 9F are views showing a method for extracting an image of an object in a grid moving method and detecting a position at which the amount of information existing in a unit region is a minimum number according to the present invention;
Fig. 10 is a flow chart showing another method for detecting a position at which the amount of information is reduced in a grid moving method according to a second embodiment of the present invention;
Fig. 11 is a flow chart showing another method for detecting a position at which the amount of information is reduced in a grid moving method according to a third embodiment of the present invention; Figs. 12A and 12B are views showing a method for sequentially outputting an image of a unit region by moving an X-axis and Y-axis grid start positions in a zig-zag manner in accordance with a signal flow of Fig. 11 according to the third embodiment of the present invention;
Fig. 13 is a view showing another method for detecting a position at which the amount of information is reduced in a grid moving method according to a fourth embodiment of the present invention;
Fig. 14 is a view showing another method for detecting a position at which the amount of information is reduced in a grid moving method according to a fifth embodiment of the present invention; Figs. 15 and 16 show tables showing a comparative result value of the conventional method and of the methods of the various embodiments accrding to the present invention, of which:
Fig. 15 is a table showing a comparative result value after moving a grid start potion in accordance with an image position of a young woman; and
Fig. 16 is a table showing a comparative result value after moving a grid in accordance with an image position of an old woman;
Fig. 17 is a block diagram showing a motion estimation/compaction apparatus according to the first embodiment of the present invention; Figs. 18A through 18C are views showing an image which is reformed by detecting a position at which the amount of information is reduced from an image of a moving compensable object in a motion estimation method according to the present invention;
Fig. 19 is a block diagram showing another motion estimation/compaction apparatus according to the second embodiment of the present invention; and Fig. 20 is a view showing an image frame so as to explain a time lapse-based variation according to the present invention.
MODES FOR CARRYING OUTTHE INVENTION
A grid moving method for minimizing image information of an object and an apparatus using the grid moving method and a compaction/motion estimation method using the grid moving method and an apparatus therefor will now be described with reference to Figs. 4 through 20.
As illustrated in Fig.4, reference numeral 31 denotes an address generation controller for moving an address start position at which an address is generated within the range of the X-axis and Y-axis by a predetermined distance, and reference numeral 33 denotes an address generator for generating X-axis and Y-axis addresses so that the image of an object can be segmented to a unit region and sequentially outputted in accordance with an address start position which the address generation controller 31 generates, and reference numeral 35 denotes a memory for sequentially storing an image of an object having predetermined shape information and segmenting and outputting a unit region in accordance with an address which the address generator 33 generates. As shown in Fig. 5, in the address generator 33, an X-axis range determination unit 331 determines the range of the X-axis of a unit region in accordance with the size information of the inputted image, and the Y-axis range determination unit 333 determines the range of the Y-axis of a unit region in accordance with the size information of the inputted image.
Here, when the ranges of the X-axis and Y-axis are identical, the size of the unit region can be determined using either the range of the X-axis or the range of the Y-axis.
The region address generator 335 of the address generator 33 judges the ranges of the X-axis and Y-axis of the unit region which is determined by the X-axis range determination unit 331 and the Y-axis range determination unit 333, segments the X-axis and Y-axis with respect to the unit region of the image of the object stored in the memory 35 based on the address start position outputted from the address generation controller 31 , and sequentially outputs the X-axis and Y-axis addresses of the segmented unit region. The memory 35 separates the image stored therein by one unit region from another unit region in accordance with X-axis and Y-xis addresses which are sequentially generated by the region address generator 335. For example, as shown in Fig. 6, the image of the object of the unit region is sequentially outputted, and then the image of the object of the next unit region is sequentially outputted.
In the drawings.as shown in Fig.4, reference numeral 37 denotes a region number counter for counting the number of unit regions in which the image of the object among the signals outputted from the memory 35 exists. As shown in Fig. 7, in the region number counter 37, a unit region counter 371 counts the clock signal and segments the unit region. A judging unit 373 separates the unit region of the outputted object in accordance with the output signal of the unit region counter 371 and judges as to whether the image of the object exists. An adder 375 adds the judgment signal of the judging unit 373 and then outputs the number of unit regions in which the image of the object exists. In the drawings.as shown in Fig.4, reference numeral 39 denotes a minimum unit region grid selector for selecting the X-axis and Y-axis grid positions at which a minimum number of the regions is counted by the region number counter 37 .
The minimum unit region grid selector 39 stores the count value when the counting of region number counter 37 is completed, and controls the address generation controller 31 to move the start positions of the X-axis and Y-axis addresses within the range of the X-axis and Y-axis of the unit region by a predetermined distance.
Namely, the minimum unit region grid selector 39 moves the start positions of the X-axis and Y-axis addresses and controls the address generation controller 31 when the counting of the number of the unit regions is completed, and then moves the start positions of the X-axis and Y-axis addresses. The above routines are repeated.
In addition, when the counting of the number of the unit regions in which the image of the object exists by moving the start positions of the X-axis and Y-axis addresses within the ranges of the X-axis and Y-axis is completed, the minimum unit region grid selector 39 judges the start positions of the X-axis and Y-axis at which the unit region of the minimum number among the number of the counted unit regions and deteimines and outputs the start positions of the judged X-axis and Y-axis addresses as the positions at which the amount of information is reduced.
Fig. 8 shows a flow chart of the method for detecting a position at which the amount of information is reduced in accordance with a first embodiment of the grid moving method.
In a step S11, X-axis and Y-axis grid start positions XM and YN are initialized so as to detect a position at which the amount of information is reduced in which the number of unit regions in which the image of the object having predetermined shape information exists.
The initial values of the X-axis and Y-axis grid start positions XM and YN are given as XM= 0, and YN = 0 which are referred to as the initial position of the minimum unit region positioned at the leftmost and uppermost portion among the unit regions with respect to the image of the extracted object.
Figs. 9A through 9F illustrate a method for adjusting the grid position so that the image of one object "a" in one image frame containing the images of a plurality of objects "a, b, c, and d" each having predetermined shape information is extracted and then exist in one image frame.
As shown in Fig. 9A, the grid is indicated with respect to one image frame containing the images of the plurality of the objects "a, b, c, and d" as shown in Fig. 9B and then the image of the objects is extracted. There are two methods for extracting the image of the objects.
One method is directed to separating and extracting so that each separate image of the objects "a, b, c, and d" can exist within one egion, and the other method is directed to extracting the image of the objects "a, b, c, and d" so that more than two images can exist within one region. As shown in Fig. 9C, the image of one object "a" is extracted, and then the image of the extracted object "a" can exists within the minimum unit region 33 so as to detect a position at which the amount of information is reduced. In addition, the entire grid with respect to the image of the extracted object "a" is called a first extraction grid.
The first extraction grid with respect to the image of the extracted object "a" is segmented into a plurality of unit regions as shown in Fig. 9C.
When the image of the object "a" is extracted, in a step S12, as shown in Fig.8,the number of the unit regions in which the image of theobject "a" exists based on the X-axis and Y-axis grid start position "A" which is initialized in the step S11 is counted. As shown in Fig. 9C, the unit region positioned at the left side and uppermost portion of the first extraction grid refers to the minimum unit region 33.
In addition, on the assumption that a unit region which is segmented by the grid has an 8 X 8 number of pixels in the X-axis and Y-axis directions, there are 64 pixels in one unit region.
In the minimum unit region, the initial positions of XM = 0, and YN = 0 refers to the start position "A" of the first extraction grid. As shown in Fig. 9C, the image of the object "a" exists in 14 unit regions.
When counting of the number of the unit regions in which the image of the object "aM exists is completed, the number of the unit regions which is counted in a step S13 is stored.
In a step S14, the grid is reformed by moving the X-axis grid start position XM along the X-axis in the minimum unit region by a predetermined distance MK", which is same as a length of a pixel along the X-axis, and in a step S15, it is judged as to whether the X-axis grid start position is moved along the X-axis M-times.
Namely, it is judged as to whether the X-axis grid start position XM is moved along the X-axis by more than the size of the unit region.
In the step S15, when the X-axis grid start position XM is not moved along the X-axis M-times, the steps S12 through S15 are performed, and then the X-axis grid start position XM is moved along the X-axis by a distance "K", and the number of the unit regions in which the image of the object "a" exists is counted and then stored. The above routines are repeated.
In the step S15, the entire X-axis grid start position XM is moved M-times, the X-axis grid start position XM in which the unit region having the minimum number in a step S16 is counted is determined as an optimum X-axis grid start position XoM.
In a step S17, the grid is reformed with the optimum X-axis grid start position XoM and the Y-axis grid start position "YN = 0", and the number of the unit regions in which the image of the object "a" exists. The counting of the number of the unit regions is completed in the step S17, in a step S18, the number of the counted unit regions are stored.
In a step S19, the Y-axis grid start position YN is moved along the Y-axis by a predetermined distance "L" which is same as a length of a pixel along the Y-axis, and in a step S20, it is judged as to whether the Y-axis grid start position YN is moved along the Y-axis N-times.
Namely, it is judged as to whether the Y-axis grid start position YN is moved along the Y-axis by more than the size of the unit region.
As a result of the step S20, when the Y-axis grid start position YN is not moved along the Y-axis N-times by a predetermined distance "L", the steps S17 through S20 are performed. The Y-axis grid start position
YN is moved based on the optimum X-axis grid start position XoM by a predetermined distance "L", and the number of the unit regions in which the image of the object "a" exists is counted and then stored. The above routines are repeated.
In the step S20, when the Y-axis grid start position YN is moved by a predetermined distance "L" N-times, the Y-axis grid start position XM in which the unit region having the minimum number is counted is determined as the optimum Y-axis grid start position YoN, and the optimum X-axis grid start position XoM and the optimum Y-axis grid start position YoN which are determined in the steps S16 and S21 are outputted to define the position "B" at which the amount of information is reduced.as shown in Fig.9D. The positionΕ" is an intersecting position of the optimum X-axis grid start position XoM and the optimum Y-axis grid start position YoN .
Namely, the first embodiment of the present invention of Fig. 8 is directed to the X-axis grid start position XM, at which the minimum number of the unit regions, in which the image of the object exists, is counted by moving the entire grid along the X-axis by a predetermined distance "K" M-times, is counted, as the optimum X-axis grid start position XoM. The Y-axis grid start position YN at which the unit regions having the minimum number in which the image of the object exists is counted by moving the entire grid along the Y-axis by a predetermined distance "L" N-times based on the determined optimum X-axis grid start position XoM is determined as the optimum Y-axis grid start position YoN. The determined X-axis grid start position XoM and the optimum Y-axis grid start position YoN are outputted to define the position at which the amount of information is reduced.
Therefore, the first embodiment of the present invention is directed to moving the entire grid along the X-axis M-times and moving the entire grid along the Y-axis N-times. Namely, the X-axis and Y-axis grid start positions XM and YN are moved M + N times, and then a grid start position "B"defined by the optimum X-axis and Y-axis grid start positions XoM and YoN is detected and outputted. Fig. 10 shows a method for detecting a position at which the amount of information is reduced in the compaction method according to a second embodiment of the present invention.
In a step S31, the X-axis and Y-axis grid start positions XM and YN are initialized as XM = 0, and YN = 0 so as to detect the optimum X-axis and Y-axis grid start positions XoM and YoN in which the number of the unit regions in which the image of the object exists becomes minimum.
In a step S32, the number of the unit regions in which the image of the object exists is counted based on the X-axis and Y-axis grid start positions XM = 0, and YN = 0 which are initialized in the step S31, and in a step S33, the number of the unit regions counted in the preceding step is stored.
In a step S34, the X-axis grid start position XM is moved along the
X-axis by a predetermined distance "K", and in a step S35, the X-axis grid start position XM is moved along the X-axis more than M-times, and then it is judged as to whether it is moved more than the size of a unit region.
As a result, when the X-axis grid start position XM is not moved along the X-axis more than M-times in the steps S35, the steps S32 through S35 are performed. The X-axis grid start position XM is moved along the X-axis by a predetermined distance "K", and the number of the unit regions in which the image of the object exists is counted and stored. The above routines are repeated. As a result, when the X-axis grid start position XM is moved along the X-axis more than M-times in the step S35, in a step S36, the Y-axis grid start position YN is moved along the Y-axis by a predetermined distance
III II
In a step S37, it is judged as to whether the Y-axis grid start position YN is moved along the Y-axis by a predetermined distance "L" more than N-times.
As a result, when the Y-axis grid start position YN is not moved along the Y-axis more than N-times in the step S37, the steps S32 through S37 are performed. The Y-axis grid is moved along the Y-axis by a predetermined distance "L", and then the X-axis grid is moved along the X-axis within the range of the unit region by a predetermined distance "K", and the number of the unit regions in which the image of the object exists is counted and stored. The above routines are repeated. As a result, when the Y-axis grid start position YN is moved along the Y-axis more than N-times in the step S37, the X-axis grid start position XM and the Y-axis grid start position YN at which the unit region having the minimum number in which the image of the object exists is counted are determined as an optimum X-axis grid start position XoM and an optimum Y-axis grid start position YoN, and then the determined optimum X-axis and Y-axis grid start positions XoM and YoN are outputted to define the position at which the amount of information is reduced.
The second embodiment of the present invention of Fig. 10 is directed to moving the entire grid along the X-axis by a predetermined distance "K" M-times and to moving the entire grid along the Y-axis by a predetermined distance "L" M-times, and further moving the entire grid along the X-axis by a predetermined distance "K" M-times and to moving the entire grid along the X-axis by a predetermined distance "L" M-times so as to count the number of the unit regions in which the image of the object exists. In addition, the X-axis and Y-axis grid start positions XM and YN at which the unit region having the minimum number is counted are determined as the optimum X-axis and Y-axis grid positions to define a position at which the amount of information is reduced and then are outputted.
Therefore, the second embodiment of the present invention of Fig. 10 is directed to moving the X-axis and Y-axis grid start position XM and YN by a predetermined distance "K" and "L" M x N times and outputting the X-axis and Y-axis grid start positions XM and YN, at which the minimum number of unit regions is counted.
The embodiments of the present invention of Figs. 8 and 10 are directed to moving the grid start position along the X-axis by a predetermined distance "K", to moving the grid start position "A" along the Y-axis by a predetermined distance "L" and to detecting the optimum X-axis and Y-axis grid start positions XoM and YoN.
However, the embodiment of the present invention of Figs. 8 and 10 are further directed to gradually moving to the Y-axis by a predetermined distance "L" and to the X-axis by a predetermined distance "K" and detecting the optimum X-axis and Y-axis grid start positions XoM and YoN and outputting the position at which the amount of information is reduced. With regard to the first and second embodedments of the grid moving methods, as shown in Figs. 8 and 10, a way of determining the grid start position of X-axis first followed by determining the grid start position of Y-axis is described for illustrative purpose.
As indicated in the brackets of Figs. 8 and 10, the grid moving methods can also be processed by determining the grid start position of Y-axis first followed by determing the grid start position of X-axis. Fig. 11 shows a method for detecting a position at which the amount of information is reduced in the grid moving method according to a third embodiment of the present invention.
In a step S41, the X-axis and Y-axis grid start positions XM and YN are initialized to 0 so as to detect the optimum X-axis and Y-axis grid start positions XoM and YoN at which the number of the unit regions in which the image of the object exists becomes minimum.
In a step S42, the number of the unit regions in which the image of the object exists is counted from the initialized X-axis and Y-axis grid start positions XM and YN, and in a step S43, the number of the counted unit regions is stored.
In a step S44, the X-axis grid start position XM is moved along the X-axis M-times, and it is judged as to whether the Y- axis grid start position YN is moved along the Y-axis N-times.
As a result, when the X-axis grid start position XM is not moved along the X-axis more than M-times or when the Y-axis grid start position YN is not moved along the Y-axis more than N-times, in step S45, the X-axis and Y-axis grid start positions XM and YN are moved within the unit region in a zig-zag manner by predetermined distances "K" and "L", and then the steps S42 through S45 are performed. Thereafter, the number of the unit regions in which the image of the object exists is counted and stored. The routines of moving the X-axis and Y-axis grid start positions XM and YN are repeated. Here, there are two methods of moving the X-axis and Y-axis grid start positions XM and YN in a zig-zag manner by predetermined distances "K" and "L". For example, as shown in Figs. 12A and 12B, there is shown a first method for moving to the direction indicated by the arrow and a second method for moving to the direction indicated by the arrow.
In the step S44, when the X-axis grid start position XM is moved along the X-axis M-times, and the Y-axis grid start position YN is moved along the Y-axis N-times, in a step S46, the X-axis grid start position XM and the Y-axis grid start position YN at which the unit regions of the minimum number are counted are determined as the optimum X-axis grid start position XoM and the optimum Y-axis grid start position YoN, and then the determined optimum X-axis grid start position XoM and the determined optimum Y-axis grid start position YoN are outputted to define the positoin "BM.
Namely, another embodiment of the present invention of Fig. 11 is directed to moving the entire grid within the range of the X-axis and Y-axis of the unit region in a zig-zag manner and to counting the number of the unit regions in which the image of the object exists. Therefore, the third embodiment of the present invention of Fig. 11 is directed to moving the X-axis and Y-axis grid start positions XM and YN by predetermined distances "K" and "L" M x N times and determining the X-axis and Y-axis grid start positions XM and YN, at which a minimum number of unit regions is counted, as the position B",as shown in Fig.9D, at which the amount of information is reduced.
Fig. 9D shows an example for reforming the grid to the position at which the amount of information is reduced which is determined in accordance with the embodiments of the present invention of Figs. 8, 10, and 11.
Here, the optimum X-axis grid start position XoM which is referred to as the position at which the amount of information is reduced existing in the unit region of the minimum number of the image of the object "a" is five (M=5), and the optimum Y-axis grid start position YoN is six (N=6).
In addition, as a result that the optimum X-axis and Y-axis grid start positions XoM and YoN are moved along the image position of the object "a", the number of the unit regions in which the image of the object "a" exists is reduced from 14 to 7.
In the above embodiments, the square-shaped unit region or the rectangular-shaped unit region which are defined by the X-axis and Y-axis are described.
Various forms of the unit region may be used so as to implement the objects of the present invention.
For example, as shown in Figs. 2A through 2F, the 45° rotated square, the hexagonal- shaped form or the like may be used. In addition, more than two different shapes which are capable of dividing the image frame into a predetermined shape at a regular interval may be combined so as to form the unit regions.
Fig. 13 shows a flow chart of a method for detecting the position at which the amount of informationis reduced in a grid moving method in accordance with a fourth embodiment of the present invention.
In a step S51, the X-axis and Y-axis grid start positions XM and YN are initialized as 0 so as to detect the position at which the number of the unit regions in which the image of the object exists, becomes minimum.
In a step S52, the number of the unit regions in which the image of the object exists from the X-axis and Y-axis grid start positions XM = 0 and YN = 0 which are initialized in the step S51 is counted, and in a step S53, the number of the unit regions counted in the step S52 is determined.
In a step S54, the Y-axis grid start position YN is moved along the Y-axis by a predetermined distance "L", and in a step S55, it is judged as to whether the Y-axis grid start position YN is moved along the Y-axis more than N-times.
As a result, when the Y-axis grid start position YN is not moved along the Y-axis more than N-times, the steps S52 through S55 are performed. The routines that the Y-axis grid start position YN is moved along the Y-axis by a predetermined distance "L", and the number of the unit regions in which the image of the object exists is counted and stored are repeated.
As a result, when the Y-axis grid start position YN is moved along the Y-axis by a predetermined distance "L" more than N-times in the step
S55, in step S56, the Y-axis grid start position YN at which a minimum number of unit regions exists is determined as the optimum Y-axis grid start position YoN.
When the optimum Y-axis grid start position YoN at which the minimum number of unit regions exists is determined in the step S56, in step S57, the grid is reformed based on the determined optimum Y-axis grid start position YoN, and in step S58, the unit region of the current X-axis row in which the image of the object exists is counted.
Namely, as shown in Fig. 9E, the unit regions in which the image of the object among the unit regions of the first row of the X(1) row in the X-axis direction exists is counted.
In step S59, the number of the counted unit regions is stored. In step S60, the grid start position XM of the X(1) row is moved along the X-axis by a predetermined distance "K", and in step S61 it is judged as to whether the X-axis grid start position XM of the X(1) row is moved along the X-axis by a predetermined distance "K" more than M-times.
As a result, when the X-axis grid start position XM of the X(1) row is not moved along the X-axis more than M-times in the step S61, the steps S58 through S61 are performed, and the routines that the X-axis grid start position XM of the X(1) row is moved along the X-axis by a predetermined distance "K", and the unit region in which the image of the object among the unit regions of the X(1) row exists is counted and stored are repeated.
In the step S61, when the grid start position XM of the X(1) row is moved by a predetermined distance M-times, in a step S62, the grid start position XM of the X(1) row at which the minimum number of unit regions among the currently counted unit regions is counted is determined as the optimum X(1) row grid start position X1M.
In step S63, it is judged as to whether the row refers to the last row of the X-axis, and when the row is not referred to as the last row of the X-axis, in a step S64, the row is moved along the next row, and then the steps S58 through S64 are performed.
The above routines are repeatedly performed, and the row of the X-axis is sequentially moved along X(1), X(2), X(3), X(4), and X(5), and the grid start position XM of the X(1), X(2), X(3), X(4), and X(5) rows at which the minimum number of unit regions in which the image of the object exists is counted are determined as the grid start positions (X1M, X2M, ...) of the optimum X(1), X(2), X(3), X(4), and X(5) rows. In the step S64, when the row refers to the last row of the X-axis, the optimum Y-axis grid start position YoN, the optimum grid start positions (X1M, X2M, ...) of the X(1), X(2), X(3), X(4), and X(5) are outputted as positions at which the amount of the information is reduced. Fig.14 shows the flow chart of a method for detecting the positions at which the amount of information is reduced in the grid moving method in accordance with a fifth embodiment of the present invention.
In step S71, the X-axis and Y-axis grid start positions XM and YN are initialized as 0 so as to detect the positions at which the number of the unit regions in which the image of the object exists becomes minimum.
In step S72, the unit region in which the image of the object exists among the unit regions of the current X-axis row is counted.
Namely, the unit region in which the image of the object exists among the unit regions of the X(1) row is detected and counted.
When the count of the unit regions is completed in the step S72, in step S73, the number of counted unit regions is stored.
In step S74, the X-axis grid start position XM of the X(1) row is moved along the X-axis by a predetermined distance "K", and in step S75, it is judged as to whether the X-axis grid start position XM of the X(1) row is moved by a predetermined distance "K" more than M-times.
In the step S75, when the grid start position XM of the X(1) is not moved by a predetermined distance "K" more than M-times, the steps S72 through S75 are performed. The routine that the X-axis grid start position
XM of the X(1) row iS moved by a predetermined distance "K", and the number of the unit regions in which the image of the object exists is counted is repeatedly performed.
In the step S75, when the X-axis grid start position XM of the X(1) row is moved along the X-axis by a predetermined distance "K" more than M-times, the grid start position XM of the X(1) row at which the minimum number of unit regions among the currently counted unit regions of the X(1) is counted is determined as the optimum X(1) row grid start position X1M. In step S77, it is judged as to whether the row refers to the last row of the A-axis, and when the row is not the last row of the X-axis, in step S78, the row is sequentially moved along the next row of the X-axis, namely, to the X(2), X(3), X(4), and X(5) rows, and the steps S72 through S78 are performed, and the routines that the grid start position XM of the X(2), X(3), X(4), and X(5) at which the minimum number of unit regions in which the image of the object exists is counted is determined as the grid start positions (X1M, X2M, ...) of the optimum X(2), X(3), X(4), and X(5) rows are repeatedly performed.
In the step S77, when the row refers to the last row of the X-axis, in step S79, it is judged as to whether the Y-axis start position YN is moved along the Y-axis by a predetermined distance "L" more than N-times. In the step S79, when the Y-axis grid start position YN is not moved along the Y-axis by a predetermined distance "L" more than N-times, in step S80, the Y-axis grid start position YN is moved along the Y-axis by a predetermined distance "L", and the steps S72 through S80 are repeatedly performed. Namely, the number of the unit regions in which the image of the object exists is counted by moving the Y-axis grid start position YN along the Y-axis by a predetermined distance L" and by sequentially moving the grid start position XM of the X(1), X(2), X(3), X(4), and X(5) by a predetermined distance "K" from the position to which the Y-axis grid start position YN is moved. In addition, the grid start position XM of the X(1), X(2), X(3), X(4), and X(5) rows at which the minimum number of unit regions is counted is sequentially determined as the grid start positions (X1M, X2M, ...) of the optimum X(1), X(2), X(3), X(4), and X(5) rows.
In the step S80, when the Y-axis grid start position YN is moved by a predetermined distance "L" more than N-times, in step S81, the numbers of the unit regions which are counted at the grid start positions (X1M, X2M, ...) of the optimum X(1), X(2), X(3), X(4), and X(5) rows which are determined at the position to which the Y-axis grid start position YN is moved are all summed.
In addition, in step S82, as a result of the sum, the position at which the minimum number of unit regions is counted is judged as the Y-axis grid start position YN, and the judged Y-axis grid start position YN is determined as the optimum Y-axis grid start position YoN. The grid start positions (X1M, X2M, ...) of the X(1), X(2), X(3), X(4) and X(5) rows at which the minimum number of unit regions of the minimum number is counted is judged as the optimum Y-axis grid start position YoN are determined as the grid start positions(X1M, X2M, ...) of the optimum X(1), X(2), X(3), X(4) and X(5) rows. In addition, the determined optimum Y-axis grid start position YoN and the grid start positions (X1M.X2M, ...) of the X(1), X(2), X(3), X(4), and X(5) are outputted as the position at which the amount of information is reduced. The result of the reformation of the grid in accordance with the position at which the amount of information is reduced which is obtained by the embodiments of Figs. 13 and 14 are shown in Fig. 9E. Here, in the fourth and fifth embodiments of Figs. 13 and 14, an example of moving the Y-axis grid start position YN and then the X-axis grid start position XM so as to detect the optimum grid start position is described. Namely, the present invention is directed to detecting the position at which the amount of informationis reduced in which the image of the object exists in the minimum number of unit regions by changing the X-axis grid start position XM and the Y-axis grid start position YN as shown in Figs. 13 and 14. Similarly.the result of the reformation of the grid to the position at which the amount of informationis reduced by changing the X-axis grid start position XM and the Y-axis grid start positions in each column(Y1N,Y2N,Y3n,Y4N and Y5N) is shown in Fig. 9F.
With regard to the fourth and fifth embodiments of the grid moving methods, as shown in Figs. 13 and 14, a way of determining the grid start position of each row of X-axis after determining the optimum Y-axis grid start position is described for illustrative purpose.
As indicated in the brackets of Figs. 13 and 14, the grid moving methods can also be processed by determining the grid start position of each column after determining the optimum X-axis grid start position.
In addition, in the second and third embodiments of the present invention of Figs. 10 and 11, the square-shaped unit region and the rectangular-shaped unit region have been explained. So as to implement the embodiments of the present invention, the unit region may be formed in various shapes. The unit regions may be moved by separating the row of the X-axis or the row of the Y-axis. In addition, the unit region may be formed in a 45° rotated square shape and then the unit region is segmented by a slant grid. The unit region may be also formed of movable slant grid.
For example, when detecting the position at which the amount of informationis reducedn by moving the unit regions of a row to the X-axis, as shown in Fig. 2A, a unit region may be formed as a triangle using two slant grids 15 and 17 opposed to the X-axis grid 11. When detecting the position at which the amount of informationis reduced by moving the unit region of the row of the Y-axis to the X-axis, as shown in
Fig. 2B, a unit region may be formed as a triangle using two slant grids 15 and 17 opposed to the Y-axis grid 13.
In addition, when detecting the position at which the amount of information is reduced by moving a unit region in an inclination manner, as shown in Fig. 2C, the unit region may be formed as a 45° rotated square using two opposed slant grids 15 and 17. In the present invention, a method of extracting the image of the object "a" and positioning it into the minimum number of unit regions has been described.
So as to implement the embodiment of the present invention, more than two images of the objects "a, b, c, and d" are selectively extracted, and then the position at which the amount of information is reduced existing in the minimum number of unit regions may be detected. In addition, the predetermined distances K and L of the X-axis and
Y-axis grid start positions XM and YN have the reference of the number of the pixels existing in the range of the unit region. For example, the X-axis and Y-axis grid start positions XM and YN may be moved by the length of the unit pixel within the range of the
X-axis and Y-axis of the unit region. However, since the information with respect to the chrominance signal from the video signal is referred to 1/2 of the information of the luminance signal, the moving distances K and L" of the X-axis and Y-axis grid start positionin XM and YN consideration of the information with respect to the chrominance signal and the luminance signal are preferably referred to as the length of two pixels.
In addition, in this embodiment of the present invention.an example that the position at which the image of the object exists in the minimum number of unit regions has been described as one position; but a plurality positions at which the amount of information is reduced may be considered.
Therefore, in the present invention, when a plurality of positions at which the amount of information is reduced are considered, the unit region is divided into sub-regions having the size of (M/2) X (N/2), and the grid start positions XM and YN are moved by predetermined distances "K and L" within the range of the X-axis and Y-axis of the divided sub-regions, and the X-axis and Y-axis grid start positions at which the image of the object exists in the minimum number of unit regions is detected and outputted as the position at which the amount of information is reduced.
Namely, on assumption that the size of the unit region is formed of
16 X 16 pixels, the sub-region has 8 X 8 pixels. In addition, the grid start positions XM and YN are moved by predetermined lengths "K and L" within the number of pixels of the sub-region which is divided into 8 X 8 pixels.
Thereafter, the X-axis and Y-axis grid start positions at which the minimum number of unit regions in which the image of the object exists is detected, and then the X-axis and Y-axis grid start positions are outputted as the positions at which the amount of information is reduced .
In addition, when a unit region is divided into the sub-regions, a plurality of positions at which the amount of information is reduced for counting the minimum number of unit regions may be generated.
Therefore, in the present invention, when a unit region is divided into the sub-regions, and a plurality of the optimum X-axis and Y-axis grid start positions XoM and YoN are generated, a proper one among them should be selected.
At this time, as one position to be selected becomes closer to the initial X-axis and Y-axis grid start positions XM = 0 and YN = 0, the value of the motion vector becomes smaller, and the information amount is decreased. In addition, when estimating the motion, since the estimation error occurrence rate becomes lower, the X-axis and Y-axis grid start positions in which the distance is nearest therebetween based on the initial grid start positions XM = 0 and YN = 0 are determined as the positions at which the amount of information is reduced.
The results of Figs. 15 and 16 were obtained by adjusting the positions of the X-axis and y-axis grids with respect to the image of the object and coding using a computer simulation.
The image used for the object was that of a young woman and an old woman, and the number of the image frames was 50, respectively.
Here, the compaction with respect to the image of the object in the conventional art refers to a shape adaptive discrete cosine transform method. Here, a fixed block grid(FBG) shape adaptive discrete cosine transform (SADCT) refers to a method of adjusting the position of the X-axis grid and Y-axis grid and coding in accordance with the first.second and third embodiments of Figs. 8, 10, and 11.
In the fourth embodiment of Fig. 13, a method of adjusting and coding the position of the X-axis grid and Y-axis grid in accordance with the X-axis grid and Y-axis grid refers to a variable block grid (VBG) shape adaptive discrete cosine transform (SADCT) 1-X method and a variable block grid (VBG) shape adaptive discrete cosine transform (SADCT) 1-Y method.
In the fifth embodiment of Fig. 14, a method of adjusting and coding the position of the X-axis grid and Y-axis grid in accordance with the X-axis grid and Y-axis grids refers to a variable block grid (VBG) shape adaptive discrete cosine transform (SADCT) 2-X method and a variable block grid (VBG) shape adaptive discrete cosine transform (SADCT) 2-Y method. In the fifth embodiment of Fig. 14, a method of separating and extracting the image of the object and then coding in accordance with the X-axis grid is referred to a variable block grid (VBG) shape adaptive discrete cosine transform (SADCT) 2-X method (by the object).
In this embodiment, an image of the object between N-1 and N frames of the original image is extracted, and the result between the conventional compaction method which is directed to performing the shape adaptive discrete cosine transform (SADCT) without varying the position of the X-axis grid and Y-axis grid from the extracted image and the compaction method of the present invention was analyzed.
The comparison method was conducted by judging how much transmission bit rate occurs as the occurrence bits per pixel(BPP) per frame and the number of regions per frame when the pixel of the object is identical with the peak signal to noise ratio(PSNR) which is the objective picture quality evaluation reference.
Fig. 15 shows the average result which is obtained by an experiment of 50 frames with respect to the image of the young woman. As shown in the table of Fig. 15, in the conventional method of performing the SADCT without varying the X-axis grid and Y-axis grid, the average peak signal to noise ratio(PSNR) value was 36.46db, the number of the average occurrence bits was 751, and the number of the average BPP was 1.21, and the number of the average unit regions was 20.71. Meanwhile, after reforming the grid using the X-axis and Y-axis which is moved so that the image of the object can exist in the minimum number of unit regions, the FBG-SADCT was performed. As a result, the average PSNR was 36.37db, and the number of the average occurrence bits was 719, and the average BPP was 1.16, and the number of the average unit regions was 18.65. In a state that the objective PSNR is similar, the FBG- SADCT of the present invention had the average number of the occurrence bits reduced by 32 bits, the BPP was reduced by 0.05, and the number of the average unit regions was reduced by 2.06.
In addition, when moving the X-axis grid and Y-axis grid after separating and dividing the image of the object.a better result appeared according to the result of the embodiments of the present invention.
As a result of the VBG-SADCT 2-X method with respect to the image of the object, the average PSNR value was 36.3db, and the number of the average occurrence bits was 694, and the average BPP was 1.12, and the number of the average regions was 16.82.
Therefore, in the present invention, it is possible to reduce the average occurrence bit rate by 57 bits performing the VBG-SADCT by the image of the object after adjusting the position of the grid in accordance with the position in which the image of the object exists compared to the conventional method which is directed to performing the SADCT without moving the position of the grid in accordance with the position at which the image of the object exists. In addition, in the present invention, the average BPP is reduced by 0.09, and the number of the average unit regions is reduced by 3.89.
Fig. 16 shows the result which is obtained by performing the SADCT without varying the grid in the conventional method with respect to the image of the object among the image of the old woman and the result which is obtained by performing the FBG-SADCT and the VBG-SADCT after varying the position of the grid in the compaction method according to the present invention. As a result, it is possible to reduce the amount of bits by about
8-10% by performing the VBG-SADCT 2-X rather than by performing the SADCT by the image of the object.
Meanwhile, Fig. 17 shows a block diagram showing the compaction/motion estimation apparatus according to a first embodiment of the present invention, which is directed to detecting the grid moving position and the position at which the amount of information is reduced with respect to the image of the object having predetermined shape information and estimating the compaction and motion after reforming the grid in accordance with the position at which the amount of informationis reduced.
In the drawings, reference numeral 41 denotes an image signal input unit for inputting the image of the object having predetermined shape information. Reference numeral 43 denotes a grid moving unit for moving the position of the X- axis grid and the Y-axis grid in accordance with the position in which the image of the object exists and for detecting a position at which the image of the object exists in the minimum number of unit regions.
In addition, the grid moving unit 43 having the same construction as the grid moving apparatus as shown in Fig. 4 is directed to sequentially storing the image of the object which is inputted by the image signal unit 41 , moving the grid start position with respect to the stored image of the object at a predetermined distance within the range of the X-axis and Y-axis, and separating the image of the object into a plurality of unit regions in accordance with the moved grid start position. In addition, the grid moving unit 43 is directed to judging and counting the unit regions in which the image of the object exists among the unit regions and outputting the X-axis and Y-axis grid start positions at which the minimum number of unit regions among the counted values is counted as the positions at which the amount of information is reduced. In the drawings, reference numeral 45 denotes a compaction unit for reforming the X-axis grid and the Y-axis grid in accordance with the positions at which the amount of information is reduced outputted from the grid moving unit 43 and for coding the image of the object.
The compaction unit 45 is directed to reforming the grid in accordance with the positions at which the amount of information is reduced, namely, which is referred to as the X-axis grid start position and Y-axis grid start position at which the image of the object exists in the minimum number of unit regions. Here, the positions at which the amount of information is reduced is detected by the grid moving apparatus and the grid moving method.
As a method for coding the image of the object of the unit regions using the compaction unit 45, there are many methods.
For example, the compaction is performed using SADCT, DCT, vector quantumization or the like.
In the drawings, reference numeral 47 denotes a motion estimation unit for estimating the motion of the image of the object. Here, the motion estimation unit 27 is directed to reforming the grid in accordance with the positions at which the amount of information is reduced, which is detected by the grid moving apparatus and the grid moving method, segmenting the unit regions in which the image of the object exists in the reformed grid, and generating the motion information by estimating the varied amount of the segmented unit regions.
On the assumption that the image of the object as shown in Fig. 18A is given, when indicating the grid as shown in Fig. 18B so as to estimate the motion of the image of the given object, the number of the unit regions in which the image of the object exists in each row is referred to 3, 5, 4, 4, 4, 5, 7,8, 8. Namely, the images of the motion compensable object exist in all 48 unit regions. Therefore, the present invention is directed to estimating the motion by reforming the grid from the above- mentioned object as shown in Fig. 18C in accordance with the positions at which the amount of information is reduced which are detected by the grid moving apparatus and the grid moving method. As a result of the reformation of the grid in accordance with the positions at which the amount of information is reduced, the number of the unit regions in which the image of the object exists in each row is referred to 3, 4, 4, 3, 3, 5, 7, 7, 8. Namely, it appeared that the number of the unit regions is reduced to all 44 unit regions. Therefore, it is possible to reduce the motion information amount by estimating the motion of the image of the object with respect to the reduced number of unit regions. Fig. 19 shows a compaction/motion estimation apparatus according to a second embodiment of the present invention which is directed to separating the image of the object having predetermined shape information into the motion compensable object and the motion compensable failed object, coding the separated motion compensable failed object, and estimating the motion of the motion compensable object.
Here, in the drawings, reference numeral 51 denotes an image signal input unit for inputting an image signal having predetermined shape information to be coded. Reference numeral 53 denotes an image separation unit for separating the image signal outputted from the image signal input unit 51 into the image of the background image having no motion and the image of the moving object having motion.
Here, the changed region of the image of the moving object is judged using the information varied between the previously inputted image and the currently inputted image.
Fig. 20 shows the time-based variation of the inputted image frame. As shown therein, it appeared that a plurality of intermediate image frames Bi, ..., Bj exist between the frame P of the previously inputted image and the frame P of the currently inputted image.
Therefore, for the image of the moving object, there is a forward motion estimation of estimating the frame Bi of the intermediate image from the frame P of the previous image and a backward motion estimation method of estimating the frame Bj of the intermediate image of the frame P of the current image.
The image of the moving object estimated by the image separation unit 53 is separated into moving information of the motion compensable object, shape information of the motion compensable object, signal information of the motion compensable failed object, and shape information of the motion compensable failed object.
In the drawings, reference numeral 55 denotes a first grid moving unit for detecting the positions at which the amount of information is reduced using shape information of the motion compensable failed objectseparated by the image separation unit 53.
The first grid moving unit 55 has the same construction as the grid moving unit 43 and as the grid moving apparatus of Fig. 4. The first grid moving unit 55 is directed to adjusting the grid position in accordance with the position at which the image of the motion compensable failed object separated by the image separation unit 53 and outputting the position in which the image of the motion compensable failed object exists in the minimum number of unit regions. In the drawings, reference numeral 57 denotes a compaction unit for coding a signal of the position at which the amount of information is reduced outputted from the first grid moving unit 55 and the image of the motion compensable failed object in accordance with the position at which the amount of informationis reduced detected the grid moving method.
The compaction unit 57 is directed to reforming the grid using the position at which the amount of information is reduced as a reference and coding the image of the motion compensable failed object of the unit region in which the image of the motion compensable failed object exists in the unit region of the reformed grid using the methods of SADCT, DCT, or vector quantumization. In the drawings, reference numeral 59 denotes a second grid moving unit having the same construction as the grid moving apparatus of Fig. 1 , the grid moving unit 43, and the first grid moving unit 55. The first grid moving unit 55 is directed to adjusting the position of the grid in accordance with the position at which the image of the motion compensable object separated by the image separation unit 53 and outputting the position in which the image of the motion compensable object exists in the minimum number of unit regions.
Reference numeral 61 denotes a motion estimation unit which is directed to reforming the grid in accordance with the signal of the position at which the amount of information is reduced outputted from the second grid moving unit 59 and the position at which the amount of informationis reduced detected by the grid moving method, segmenting the image of the motion compensable object into the unit regions from the reformed grid, and generating the motion information using the variation amount of the segmented unit regions.
As described above, the grid moving method for minimizing image information of an object and an apparatus using the grid moving method and the compaction/motion estimation method using the grid moving method and an apparatus thereof according to the present invention are directed to forming the minimum number of unit regions having the image of the object by adjusting the position of the grid in accordance with the image position of the object, coding the image of the object with respect to the minimum number of unit regions , and generating the motion information, thus achieving a higher compaction rate, whereby the amount of the data to be stored and transmitted can be significantly reduced. Although the preferred embodiments of the present invention have been disclosed for illustrative purposes, those skilled in the art will appreciate that various modifications, additions and substitutions are possible, without departing from the scope and spirit of the invention as described in the accompanying claims.

Claims

1. A grid moving method forminimizing image information of an object, comprising the steps of; a forming step for forming a grid over an image of an object: a segmenting step for segmenting the grid formed over the image into a plurality of unit regions; and a moving and counting step for moving the grid and counting the number of the unit region in which the image of the object exists.
2. A method of claim 1, wherein said moving and counting step further comprises the steps of: moving the grid along an X-axis and/or along a Y-axis; judging an amount of image information at each position to which the grid is moved in the moving step; and repeating the moving step and judging step until judging the amount of the image information at all positions is completed.
3. A method of claim 1, wherein said moving and counting step further comprises the steps of: a first moving and counting step for moving the grid along an
X-axis and for counting the number of the unit regions in which the image of the object exists; a first setting step for setting an optimum X-axis grid start position on the X-axis at which the number of the unit regions counted in the first moving and counting step is minimum; a second moving and counting step for moving the grid along a Y-axis and for counting the number of the unit regions in which the image of the object exists after setting the grid in the optimum X-axis grid start position set in the first setting step; a second setting step for setting a optimum Y-axis grid start position on the Y-axis at which the number of the unit regions counted in the second moving and counting step is minimum; and a determining step for determining a position at which an amount of image information is reduced, whereby the position is an intersecting position of the optimum X-axis grid start position and the optimum Y-axis grid start position set in the first and second setting steps.
4. A method of claim 1, wherein said moving and counting step further comprises the steps of: a first moving and counting step for moving the grid along a Y-axis and for counting the number of the unit regions in which the image of the object exists; a first setting step for setting an optimum Y-axis grid start position on the Y-axis at which the number of the unit regions counted in the first moving and counting step is minimum; a second moving and counting step for moving the grid along an X-axis and for counting the number of the unit regions in which the image of the object exists after setting the grid in the optimum Y-axis grid start position set in the first setting step; a second setting step for setting an optimum X-axis grid start position of the grid on the X-axis at which the number of the unit regions counted in the second moving and counting step is minimum; and a determining step for determining a position at which an amount of image information is reduced, whereby the position is an intersecting position of the X-axis optimum grid start position and the optimum Y-axis grid start position set in the first and second setting steps.
5. A method of claim 1, wherein said moving and counting step further comprises the steps of: a first moving and counting step for moving the grid along an X-axis and for counting the number of the unit regions in which the image of the object exists; a second moving and counting step for moving the grid along a Y-axis and for counting the number of the unit regions in which the image of the object exists; repeating the first and second moving and counting steps until counting the number of the unit regions at all positions is completed; a first determining step for determining an optimum X-axis grid start position and an optimum Y-axis grid start position at which the number of the unit regions is minimum; and a second determining step for determining a position which is an intersecting position of the optimum X-axis grid start position and the optimum Y-axis grid start position determined in the first determining step.
6. A method of claim 1 , wherein said moving and counting step further comprises the steps of: a first moving and counting step for moving the grid alog a Y-axis and for counting the number of the unit regions in which the image of the object exists; a second moving and counting step for moving the grid along an X-axis and for counting the number of the unit regions in which the image of the object exists; repeating the first and second moving and counting steps until counting the number of the unit regions at all positions is completed; a first determining step for determining a optimum Y-axis grid start position and an optimum X-axis grid start position at which the number of unit regions is minimum; and a second determining step for determining a position which is an intersecting position of the optimum X-axis grid start position and the optimum Y-axis grid start position determined in the first determining step.
7. A method of claim 1, wherein said moving and counting step further comprises the steps of: a first moving and counting step for moving the grid in a zig-zag manner and for counting the number of the unit regions in which the image of the object exists; and a first determining step for determining a position at which the number of the unit regions is minimum in the first moving and counting step as the position at which an amount of compaction information is reduced.
8. The method of claim 1 , wherein said moving and counting step further comprises the steps of: a moving step for moving the grid along an X-axis and a Y-axis; a judging step for judging an amount of information at each X-axis grid position and Y-axis grid position in each row and column to which the grid is moved in the moving step; and an outputting step for detecting a position at which the amount of information is reduced in the judging step.
9. The method of claim 1, wherein said moving and counting step further comprises the steps of: a first moving and counting step for moving the grid along the Y-axis and for counting the number of unit regions in which the image of the object exists; a first setting step for setting a position at which the number of unit regions counted in the first counting step is minimum as a Y-axis grid start position; a second moving and counting step for moving the grid of each row of the X-axis to the X-axis from the Y-axis position set in the first setting step and for counting the unit regions of each row of the X-axis in which the image of the object exists; a second setting step for setting positions at which the number of unit regions counted in the second counting step is minimum as a grid start position of each row of the X-axis; and a determining step for determining the Y-axis grid start position and the grid start position of each row of the X-axis set in the first and second setting steps as the position at which the amount of information is reduced.
10. The method of claim 1, wherein said moving and counting step further comprises the steps of: a first moving and counting step for moving the grid along the X-axis and for counting the number of unit regions in which the image of the object exists; a first setting step for setting a position at which the number of unit regions counted in the first counting step is minimum as an X-axis grid start position; a second moving and counting step for moving the grid of each column of the Y-axis from the X-axis set in the first setting step along the Y-axis and for counting the unit regions of each column of the Y-axis in which the image of the object is positioned; a second setting step for setting positions at which the number of unit regions counted in the second counting step is minimum as a grid start position of each column of the Y-axis; and a determining step for determining the X-axis grid start position and the grid start position of each column of the Y-axis set in the first and second setting steps as positions at which the amount of the compaction information amount and motion information is reduced.
11. The method of claim 1, wherein said moving and counting step further comprises the steps of: an X-axis grid position determining step for moving the grid of each row of the X-axis along the X-axis, and for counting the number of the unit regions in which the image of the object exists, and determining an X-axis grid start position at which the minimum number of unit regions is counted as a grid start position of each row of the X-axis; a moving step for moving the grid along the Y-axis when the counting is completed in the X-axis position determining step and for repeating the X-axis position determining step; a Y-axis grid start position determining step for summing the counted values at the grid position of each row of the X-axis determined in the X-axis position determining step when the movement of the Y-axis is completed in the moving step and for determining a Y-axis grid start position in which the summed value is minimum; and a determining step for determining the grid start position of each row of the X-axis at which the minimum number of unit regions is counted at the Y-axis grid start position determined in the Y-axis grid start positions determining step as the grid position at which the amount of compaction information and motion information is reduced.
12. The method of claim 1 , wherein said moving and counting further comprises the steps of: a Y-axis grid position setting step for moving the grid of each column of the Y-axis to the Y-axis, and for counting the number of the unit regions in which the image of the object exists, and for setting a Y-axis grid start position at which the minimum number of unit regions is counted as a grid start position of each column of the Y-axis; a moving step for moving the grid along the X-axis when the counting is completed in the Y-axis position setting step and for repeating the Y-axis position determining step; an X-axis grid position setting step for summing the counted value at the grid start position of each column of the Y-axis set in the Y-axis position setting step when the movement above the X-axis is completed in the moving step and for setting an X-axis grid start position in which the summed value is minimum as the X-axis position setting step; and a determining step for determining the grid start position of each column of the Y-axis at which the minimum number of unit regions is counted at the X-axis grid start position set in the X-axis grid position setting step as the grid start positions at which the amount of compaction information and motion information is reduced.
13. The method of claims 1, 2, or 8, wherein said method further includes a step of detecting a position at which the amount of information is minimum.
14. The method of claims 1,2,3,4,5,6,7,8,9,10, 11, or 12, wherein each of said unit regions is square-shaped or rectangular-shaped by the X-axis grid and the Y-axis grid.
15. The method of claims 1,2,3,4,5,6,7,8,9 or 11, wherein each of said unit regions is triangle-shaped and is formed by the X-axis grid and two opposed slant grids.
16. The method of claims 1,2,3,4,5,6,7,8,10,or 12, wherein each of said unit regions is triangle-shaped and is formed by the Y-axis grid and two opposed slant grids.
17. The method of claims 1,2,3,4,5,6,7,8,9,10,11, or 12, wherein each of said unit regions is formed by a 45° -rotated square.
18. The method of claims 1,2,3,4,5,6,or 7, wherein each of said unit regions is hexagonal-shaped.
19. The method of claims 1,2,3,4,5,6,or 7, wherein each of said unit regions is formed of octagonal-shaped unit regions and 45° -rotated square-shaped unit regions.
20. The method of claims 1,2,3,4,5,6,7,8,9,10,11, or 12, wherein a moving range of the grid is within the ranges of the X-axis and Y-axis of a unit region.
21. The method of claims 1,2,3,4,5,6,7,8,9,10, 11, or 12, wherein a moving range of the grid is within the ranges of the X-axis and Y-axis of a unit region and is moved by a unit pixel.
22. The method of claims 1,2l3,4,5,6l7,8,9,10,11lor 12, wherein a moving range of the grid is within ranges of the X-axis and Y-axis and is moved by two unit pixels.
23. The method of claims 3,4,5,6,7,9,10,11, or 12, wherein said method further includes a step for segmenting ranges of the X-axis and Y-axis of a unit region by 1/2 when a plurality of X-axis and Y-axis grid positions, at which a minimum number of unit regions is counted, exist, and for counting the number of the unit regions in which the image of the object exists, and determining the positions, at which the number of unit regions counted is minimum, as the positions at which the amount of information is reduced.
24. The method of claim 23, wherein said method further includes a step for segmenting the ranges of the X-axis and Y-axis of the unit region to 1/2 and for determining the X-axis and Y-axis grid positions closest from X-axis and Y-axis grid start positions as the postitions at which the amount of information is reduced when a plurality of positions, at which the minimum number of unit regions is counted, exist.
25. A compaction/motion estimation method using a grid moving method for minimizing image information of an object, comprising the steps of: a moving step for forming a grid over an image of an object having shape information, segmenting the image into a plurality of unit regions, and moving the formed grid; a judging step for judging an amount of the shape information at each position to which the grid is moved in the moving step; a detecting step for detecting a position at which the amount of the shape information is reduced; a compaction step for reforming the grid in accordance with the position detected in the detecting step and for coding the image of the object existing in unit regions of the reformed grid; and a motion estimation step for reforming the grid in accordance with the position detected in the detecting step and for estimating the motion of the image of the object in the unit regions in which the image of the object exists among the unit regions segmented by the reformed grid.
26. The method of claim 25, wherein said compaction step is directed to performing a shape adaptive discrete cosine transform
(SADCT).
27. The method of claim 25, wherein said compaction step is directed to performing a vector quantum ization.
28. The method of claim 25, wherein said compaction step is directed to performing a discrete cosine transform (DCT).
29. A compaction method using a grid moving method for minimizing image information of an an object, comprising the steps of: a moving step for forming a grid over an image of an object having shape information, segmenting the image into a plurality of unit regions, and for moving the formed grid; a judging step for judging an amount of the shape information at each position to which the grid is moved in the moving step; a detecting step for detecting a position at which the amount of the shape information is reduced in the judging step; and a compaction step for reforming the grid in accordance with the position detected in the detecting step and coding the image of the object existing in unit regions by the reformed grid.
30. The method of claim 29, wherein said compaction step is directed to performing a shape adaptive discrete cosine transform (SADCT).
31. The method of claim 29, wherein said compaction step is directed to performing a vector quantumization.
32. The method of claim 29, wherein said compaction step is directed to performing a discrete cosine transform (DCT).
33. A motion estimation method using a grid moving method for minimizing image information of an object, comprising the steps of: a moving step for forming a grid over an image of an object having shape information, segmenting the image into a plurality of unit regions, and moving the formed grid; a judging step for judging an amount of the shape information at each position to which the grid is moved in the moving step; a detecting step for detecting a position at which the amount of shape information is reduced in the judging step; and a motion estimation step for reforming the grid in accordance with the position detected in the detecting step and for estimating the motion of the image of the object in the unit regions in which the image of the object exist among the unit regions by the reformed grid.
34. A compaction/motion estimation method using a grid moving method for minimizing image information of an object, comprising the steps of: a separating step for estimating the motion of an image of an object having shape information and for separating a motion compensable failed object image and a motion compensable object image; a first moving step for forming a grid over the motion compensable failed object image separated in the separating step, segmenting the image into a plurality of unit regions, and moving the grid; a first judging step for judging an amount of shape information for the motion compensable failed object image at each position to which the grid is moved in the first moving step; a first detecting step for detecting a position at which the amount of shape information is reduced in the first judging step; a compaction step for reforming the grid in accordance with the position detected in the first detecting step and for coding the unit regions in which the image of the motion compensable failed object exists from among the unit regions which is segmented by the reformed grid; a second moving step for forming a grid over the image of the motion compensable object separated in the separating step, segmenting the image into a plurality of unit regions, and moving the grid; a second judging step for judging an amount of shape information for the motion compensable object image at each position to which the grid is moved in the second moving step; a second detecting step for detecting a position at which the amount of shape information is reduced in the second judging step; and a motion estimating step for reforming the grid in accordance with the position detected in the second detecting step and for estimating the motion of the image of the motion compensable object using the unit regions in which the image of the motion compensable object exists from among the unit regions segmented by the reformed grid.
35. The method of claim 34, wherein said separating step is directed to estimating the motion of the image of the current object from the image of the object having previous shape information.
36. The method of claim 34, wherein said separating step is directed to estimating the motion of the image of the previous object from the image of the object having current shape information.
37. The method of claim 34, wherein said compaction step is directed to performing SADCT.
38. The method of claim 34, wherein said compaction step is directed to performing vector quantumization.
39. The method of claim 34, wherein said compaction step is directed to performing DCT.
40. A compaction method using a grid using a grid moving method for minimizing image information of an object image, comprising the steps of: a separating step for estimating the motion of the image of an object having shape information and for separating the image of a motion compensable failed object; a moving step for forming a grid over the image of the motion compensable failed object separated in the separating step, segmenting the image into a plurality of unit regions, and moving the grid; a judging step for judging an amount of shape information at each position to which the grid is moved in the moving step; a detecting step for detecting a position at which an amount of the shape information is reduced in the judging step; and a compaction step for reforming the grid in accordance with the position detected in the detecting step and for coding the unit regions in which the image of the motion compensable failed object exists among the unit regions segmented by the reformed grid.
41. The method of claim 40, wherein said separating step is directed to estimating the motion of the image of the current object among the image of the object having previous shape information.
42. The method of claim 40, wherein said separating step is directed to estimating the motion of the image of the previous object from the image of the object having current shape information.
43. The method of claim 40, wherein said compaction step is directed to performing SADCT.
44. The method of claim 40, wherein said compaction step is directed to performing vector quantumization.
45. The method of claim 40, wherein said compaction step is directed to performing DCT.
46. A motion estimation method using a grid movement of an image of an object, comprising the steps of: a separating step for estimating the motion of the image of the object having shape information and for separating the image of the motion compensable object; a moving step for forming a grid over the image of the motion compensable object separated in the separating step, segmenting the image into a plurality of unit regions, and moving the grid; a judging step for judging an amount of shape information at each position to which the grid is moved in the moving step; a detecting step for detecting a position at which an amount of the shape information is reduced in the judging step; and a motion estimating step for reforming a grid in accordance with the position detected in the detecting step and for estimating the motion of the image of the motion compensable object using the unit regions in which the image of the motion compensable object exists from among the unit regions segmented by the reformed grid.
47. The method of claim 46, wherein said separating step is directed to estimating the motion of the image of the current object from the image of the object having previous shape information.
48. The method of claim 46, wherein said separating step is directed to estimating the motion of the image of the previous object from the image of the object having current shape information.
49. A grid moving apparatus for minimizing image information of an object, comprising: address generation control means for moving an address start position at which an address is generated by a predetermined distance within a predetermined region of a unit region; address generating means for separating the image of the object into unit regions in accordance with the address start position which the address generation control means outputs and for generating the address; memory means for storing the image of the object having inputted shape information and for outputting the image in accordance with an address outputted from the address generation means; region number counting means for counting the number of unit regions in which the shape information of the object exists outputted from the memory means; and minimum unit region grid selecting means for selecting an
X-axis grid start position and Y-axis grid start position at which the minimum number of unit regions is counted from among the number of the unit regions counted by the region number counting means.
50. The apparatus of claim 49, wherein said address generation means includes:
X-axis range determining means and Y-axis range determining means for determining the X-axis range of the unit region and the Y-axis range of the unit region in accordance with size information of the inputted image of the object; and region address generation means for separating the X-axis range and Y-axis range of the unit region which are determined by the X-axis range determining means and the Y-axis range determining means from the address start position outputted from the address generation control means.
51. The apparatus of claim 50, wherein said X-axis range and Y-axis range of the unit region are determined by one size determining means when the X-axis range and the Y-axis range of the unit region are identical to each other.
52. The apparatus of claim 49, wherein said region number counting means includes: region counting means for counting a clock signal and for separating the unit region; shape information existing judging means for separating the unit region of the image outputted from the memory means in accordance with an output signal of the region counting means; and region number adding means for counting a judging signal of the shape information existing judging means and for outputting the number of the unit regions in which the image of the object exists.
53. A compaction/motion estimation apparatus using a grid moving method for minimizing image information of an object, comprising: image signal input means for inputting an image of an object having shape information; grid moving means for adjusting a grid in accordance with an image position of an object having the shape information outputted from the image signal input means and for reducing the number of unit regions in which the image of the object exists; compaction means for coding the image of the object existing in the unit regions in which the number of unit regions is reduced in the grid moving means; and motion estimation means for estimating a motion of the image of the object using the motion of unit regions of which the number of unit regions is reduced.
54. The apparatus of claim 53, wherein said grid moving means includes: address generation control means for moving an address start position at which the address is generated by a predetermined distance within the size of the unit region; address generation means for separating the image of the object into unit regions in accordance with the address start position outputted from the address generation control means and for generating the address; memory means for storing the image of the object having inputted shape information and for outputting in accordance with the address which the address generation means generates; region number counting means for counting the number of the unit regions in which the shape information of the object exists outputted from the memory means; minimum unit region grid selecting means for selecting an X-axis grid start position and a Y-axis grid start position at which the minimum number of unit region is counted from among the number of the unit regions which the region number counting means counted.
55. The apparatus of claim 54, wherein said address generation means includes:
X-axis range determining means and Y-axis range determining means for determining the X-axis range of the unit region and the Y-axis range of the unit region in accordance with size information of the inputted image of the object; and region address generation means for separating the X-axis range and Y-axis range of the unit region which the X-axis range determining means and the Y-axis range determining means determined from the address start position outputted from the address generation control means and for sequentially outputting an address in accordance with the X-axis range and Y-axis range of the unit regions separated by the region address generation means.
56. The apparatus of claim 55, wherein said X-axis range and Y-axis range of the unit region are determined by one size determining means when the X-axis range and Y-axis range of the unit region are the same.
57. The apparatus of claim 54, wherein said region number counting means includes: region counting means for counting a clock signal and for separating the unit region; shape information existing judging means for separating the unit region of the image outputted from the memory means in accordance with an output signal of the region counting means and for judging as to whether there exists a shape information of the object; and region number adding means for counting a judging signal of the shape information existing judging means and for outputting the number of the unit regions in which the image of the object exists.
58. A compaction apparatus using a grid moving of an object image, comprising: image signal input means for inputting an image of an object having shape information; grid moving means for reducing the number of unit regions in which the image of an object exists by adjusting a grid in accordance with an image position of an object having the shape information outputted from image signal input means; and compaction means for coding the image of the object existing in the unit regions of which the number of unit regions is reduced by the grid moving means.
59. The apparatus of claim 58, wherein said grid moving means includes: address generation control means for moving an address start position, at which the address is outputted, by a predetermined distance within the size of the unit region; address generation means for separating the image of the object into unit regions in accordance with the address start position outputted from the address generation control means; memory means for storing the image of the object having inputted shape information and for outputting in accordance with an address outputted from the address generation means; region number counting means for counting the number of the unit regions in which the shape information of the object exists outputted from the memory means; and minimum unit region grid selecting means for selecting an X-axis grid start position and a Y-axis grid start position at which the minimum number of unit regions is counted from among the number of the unit regions counted by the region number counting means.
60. The apparatus of claim 59, wherein said address generation means includes: X-axis range determining means and Y-axis range determining means for determining the X-axis range of the unit region and the Y-axis range of the unit region in accordance with size information of the inputted image of the object; and region address generation means for separating the X-axis range and Y-axis range of the unit region determined by the X-axis range determining means and Y-axis range determining means from the address start position outputted from the address generation control means and for sequentially outputting an address in accordance with the X-axis range and Y-axis range of the unit region separated by the region address generation means.
61. The apparatus of claim 60, wherein said X-axis range Y-axis range of the unit region are determined by one size determining means when the X-axis range and Y-axis range of the unit region are the same.
62. The apparatus of claim 59, wherein said region number counting means includes: region counting means for counting a clock signal and for separating the unit region; shape information existing judging means for separating the unit region of the image outputted from the memory means in accordance with an output signal of the region counting means and for judging as to whether there exists shape information of the object; and region number adding means for counting a judging signal of the shape information existing judging means and for outputting the number of the unit regions in which the image of the object exists.
63. A motion estimation apparatus using a grid moving of an object image, comprising: image signal input maens for inputting an image of an object having shape information; grid moving means for adjusting the grid in accordance with an image position of an object having shape information outputted from image signal input means and for reducing the number of unit regions in which the image of the object exists; and motion estimation means for estimating a motion of the image of the object using the motion of the unit regions of which the number of unit regions is reduced.
64. The apparatus of claim 63, wherein said grid moving means includes: address generation control means for moving an address start position, at which the address is outputted, by a predetermined distance within the size of the unit region; address generation means for separating the image of the object into unit regions in accordance with the address start position outputted from the address generation control means; memory means for storing the image of the object having inputted shape information and for outputting in accordance with an address outputted from the address generation means; region number counting means for counting the number of the unit regions in which the shape information of the object exists outputted from the memory means; and minimum unit region grid selecting means for selecting an X-axis grid start position and a Y-axis grid start position at which the minimum number of unit region is counted from among the number of the unit regions counted by the region number counting means.
65. The apparatus of claim 64, wherein said address generation means includes:
X-axis range determining means and Y-axis range determining means for determining the X-axis range of the unit region and the Y-axis range of the unit region in accordance with size information of the inputted image of the object; and region address generation means for separating the X-axis range and Y-axis range of the unit region determined by the X-axis range determining means and Y-axis range determining means from the address start position outputted from the address generation control means and for sequentially outputting an address in accordance with the X-axis range and Y-axis range of the unit regions separated by the region address generation means.
66. The apparatus of claim 65, wherein said X-axis range and Y-axis range of the unit region are determined by one size determining means when the X-axis range and Y-axis range of the unit region are the same.
67. The apparatus of claim 64, wherein said region number counting means includes: region counting means for counting a clock signal and for separating the unit region; shape information existing judging means for separating the unit region of the image outputted from the memory means in accordance with an output signal of the region counting means and for judging as to whether there exists a shape information of the object; and region number adding means for counting a judging signal of the shape information existing judging means and for outputting the number of the unit regions in which the image of the object exists.
68. A compaction/motion estimation apparatus using a grid moving of an object image, comprising: image separating means for separating an image of a motion compensable failed object and an image of a motion compensable object in accordance with a motion from an image of an object having shape information; first grid moving means for adjusting the grid in accordance with an image position of the motion compensable failed object separated by the image separating means and for reducing the number of unit regions in which the image of the motion compensable failed object exists; compaction means for coding the image of the object existing in the unit regions of which the number of unit regions is reduced by the first grid moving means; second grid moving means for adjusting the grid in accordance with the image position of the motion compensable object separated by the image separating means and for reducing the number of unit regions in which the image of the motion compensable object exists; and motion estimation means for estimating the motion information of the motion compensable object using the unit regions in which the image of the motion compensable object exists of which the number of the unit regions is reduced.
69. The apparatus of claim 68, wherein said first and second grid moving means includes: address generation control means for moving an address start position, at which the address is outputted, by a predetermined distance within the size of the unit region; address generation means for separating the image of the object into unit regions in accordance with the address start position outputted from the address generation control means; memory means for storing the image of the object having inputted shape information and for outputting in accordance with an address outputted from the address generation means; region number counting means for counting the number of the unit regions in which the shape information of the object exists outputted from the memory means; and minimum unit region grid selecting means for selecting an X-axis grid start position and a Y-axis grid start position at which the minimum number of unit regions is counted from among the number of the unit regions counted by the region number counting means.
70. The apparatus of claim 69, wherein said address generation means includes: X-axis range determining means and Y-axis range determining means for determining the X-axis range of the unit region and the Y-axis range of the unit region in accordance with size information of the inputted image of the object ; and region address generation means for separating the X-axis range and Y-axis range of the unit region determined by the X-axis range determining means and Y-axis range determining means from the address start position outputted from the address generation control means and for sequentially outputting an address in accordance with the X-axis range and Y-axis range of the unit region separated by the address generation means.
71. The apparatus of claim 70, wherein said X-axis range and
Y-axis range of the unit region are determined by one size determining means when the X-axis range and Y-axis range of the unit region are the same.
72. The apparatus of claim 69, wherein said region number counting means includes: region counting means for counting a clock signal and for separating the unit region; shape information existing judging means for separating the unit region of the image outputted from the memory means in accordance with an output signal of the region counting means and for judging as to whether there exists a shape information of the object; and region number adding means for counting a judging signal of the shape information existing judging means and for outputting the number of the unit regions in which the image of the object exists.
73. A compaction apparatus using a grid moving of an object image, comprising: image separating means for separating an image of a motion compensable failed object in accordance with a motion of an image of an object having shape information; grid moving means for adjusting the grid in accordance with an image position of the motion compensable failed object separated by the image separating means and for reducing the number of unit regions in which the image of the motion compensable failed object exists; and compaction means for coding the image of the object existing in the unit region of which the number of the unit regions is reduced by the grid moving means.
74. The apparatus of claim 73, wherein said first and second grid moving means includes: address generation control means for moving an address start position, at which the address is outputted, by a predetermined distance within the size of the unit region; address generation means for separating the image of the object into unit regions in accordance with the address start position outputted from the address generation control means; memory means for storing the image of the object having inputted shape information and for outputting in accordance with an address outputted from the address generation means; region number counting means for counting the number of the unit regions in which the shape information of the object exists outputted from the memory means; and minimum unit region grid selecting means for selecting an X-axis grid start position and a Y-axis grid start position at which the minimum number of unit regions is counted from among the number of the unit regions counted by the region number counting means.
75. The apparatus of claim 74, wherein said address generation means includes:
X-axis range determining means and Y-axis range determining means for determining the X-axis range of the unit region and the Y-axis range of the unit region in accordance with size information of the inputted image of the object ; and region address generation means for separating the X-axis range and Y-axis range of the unit region determined by the X-axis range determining means and Y-axis range determining means from the address start position outputted from the address generation control means and for sequentially outputting an address in accordance with the X-axis range and Y-axis range of the unit region separated by the address generation means.
76. The apparatus of claim 75, wherein said X-axis range and Y-axis range of the unit region are determined by one size determining means when the X-axis range and Y-axis range of the unit region are the same.
77. The apparatus of claim 74, wherein said region number counting means includes: region counting means for counting a clock signal and for separating the unit region; shape information existing judging means for separating the unit region of the image outputted from the memory means in accordance with an output signal of the region counting means and for judging as to whether there exists a shape information of the object; and region number adding means for counting a judging signal of the shape information existing judging means and for outputting the number of the unit regions in which the image of the object exists.
78. A motion estimation apparatus using a grid moving of an object image, comprising: image separating means for separating an image of a motion compensable object in accordance with a motion of an object having shape information; grid moving means for adjusting a grid in accordance with an image position of the motion compensable object separated by the image separating means and for reducing the number of unit regions in which the image of the motion compensable object exists; and motion estimation means for estimating motion information of the motion compensable object using the unit regions in which the image of the motion compensable object exists of which the number of the unit regions is reduced by the grid moving means.
79. The apparatus of claim 78, wherein said first and second grid moving means includes: address generation control means for moving an addresss start position, at which the address is outputted, by a predetermined distance within the size of the unit region; address generation means for separating the image of the object into unit regions in accordance with the address start position outputted from the address generation control means; memory means for storing the image of the object having inputted shape information and for outputting in accordance with an address outputted from the address generation means; region number counting means for counting the number of the unit regions in which the shape information of the object exists outputted from the memory means; and minimum unit region grid selecting means for selecting an X-axis grid start position and a Y-axis grid start position at which the minimum number of unit regions is counted from among the number of the unit regions counted by the region number counting means.
80. The apparatus of claim 79, wherein said address generation means includes:
X-axis range determining means and Y-axis range determining means for determining the X-axis range of the unit region and the Y-axis range of the unit region in accordance with size information of the inputted image of the object ; and region address generation means for separating the X-axis range and Y-axis range of the unit region determined by the X-axis range determining means and Y-axis range determining means from the address start position outputted from the address generation control means and for sequentially outputting an address in accordance with the X-axis range and Y-axis range of the unit region separated by the address generation means.
81. The apparatus of claim 80, wherein said X-axis range and Y-axis range of the unit region are determined by one size determining means when the X-axis range and Y-axis range of the unit region are the same.
82. The apparatus of claim 79, wherein said region number counting means includes: region counting means for counting a clock signal and for separating the unit region; shape information existing judging means for separating the unit region of the image outputted from the memory means in accordance with an output signal of the region counting means and for judging as to whether there exists a shape information of the object; and region number adding means for counting a judging signal of the shape information existing judging means and for outputting the number of the unit regions in which the image of the object exists.
PCT/KR1996/000106 1995-07-10 1996-07-10 Grid moving method of object image and apparatus using the same and compaction/motion estimation method using the same and apparatus thereof WO1997003525A1 (en)

Priority Applications (4)

Application Number Priority Date Filing Date Title
DE1996615805 DE69615805T2 (en) 1995-07-10 1996-07-10 METHOD AND DEVICE FOR OBJECT IMAGE GRID MOVEMENT, MOTION ESTIMATION / COMPRESSION METHOD THEREFOR, AND DEVICE THEREFOR
AU63199/96A AU6319996A (en) 1995-07-10 1996-07-10 Grid moving method of object image and apparatus using the same and compaction/motion estimation method using the same and apparatus thereof
EP96922276A EP0781492B1 (en) 1995-07-10 1996-07-10 Grid moving method of object image and apparatus using the same and compaction/motion estimation method using the same and apparatus thereof
CA 2198463 CA2198463C (en) 1995-07-10 1996-07-10 Grid moving method of object image and apparatus using the same and compaction/motion estimation method using the same and apparatus thereof

Applications Claiming Priority (8)

Application Number Priority Date Filing Date Title
KR1995-20198 1995-07-10
KR1995-20199 1995-07-10
KR19950020199 1995-07-10
KR19950020198 1995-07-10
KR1019950037919A KR970025172A (en) 1995-10-26 1995-10-26 Motion estimation method and apparatus by shape adaptive object segmentation method
KR1995-37919 1995-10-26
KR1996-6465 1996-03-12
KR1019960006465A KR970009394A (en) 1995-07-10 1996-03-12 Video signal compression device and compression method

Publications (1)

Publication Number Publication Date
WO1997003525A1 true WO1997003525A1 (en) 1997-01-30

Family

ID=27483073

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/KR1996/000106 WO1997003525A1 (en) 1995-07-10 1996-07-10 Grid moving method of object image and apparatus using the same and compaction/motion estimation method using the same and apparatus thereof

Country Status (9)

Country Link
US (4) US5892849A (en)
EP (2) EP0781492B1 (en)
JP (1) JP3242409B2 (en)
AU (1) AU6319996A (en)
CA (1) CA2198463C (en)
DE (2) DE69615805T2 (en)
MY (1) MY121607A (en)
TW (1) TW320806B (en)
WO (1) WO1997003525A1 (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0809404A2 (en) * 1996-05-23 1997-11-26 Daewoo Electronics Co., Ltd Apparatus for coding an object region of a video signal by using a rearranged block-based technique
EP0809405A2 (en) * 1996-05-23 1997-11-26 Daewoo Electronics Co., Ltd Method and apparatus for determining an optimum grid for use in a block-based video signal coding system
WO2000025232A1 (en) * 1998-10-23 2000-05-04 Siemens Aktiengesellschaft Method and array for coding, decoding, and transmitting a digitized image
WO2002071759A1 (en) * 2001-03-07 2002-09-12 Nikon Corporation Image compression device, electronic camera, and image processing program
US7561748B2 (en) 2003-07-03 2009-07-14 Nikon Corporation Electronic camera

Families Citing this family (53)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
ATE224121T1 (en) * 1996-07-31 2002-09-15 Matsushita Electric Ind Co Ltd METHOD AND DEVICE FOR IMAGE DECODING
FR2752474B1 (en) * 1996-08-14 1998-12-31 Iona Donescu PROCESS FOR TRANSFORMING THE IMAGE SIGNAL ON ARBITRARY MEDIA
GB9626359D0 (en) * 1996-12-19 1997-02-05 Olivetti Telemedia Spa Method for encoding digital information
US6389175B1 (en) 1996-12-19 2002-05-14 At&T Laboratories, Limited Method for encoding digital information
US6633611B2 (en) * 1997-04-24 2003-10-14 Mitsubishi Denki Kabushiki Kaisha Method and apparatus for region-based moving image encoding and decoding
US7093192B2 (en) * 1999-07-30 2006-08-15 Microsoft Corporation Establishing and displaying dynamic grids
US6738424B1 (en) 1999-12-27 2004-05-18 Objectvideo, Inc. Scene model generation from video for use in video processing
US6813388B2 (en) * 2000-02-07 2004-11-02 Matsushita Electric Industrial Co., Ltd. Image coding method, image coding apparatus and data storage medium
JP2002190020A (en) * 2000-12-20 2002-07-05 Monolith Co Ltd Method and device for image effect
US6450611B1 (en) 2001-02-14 2002-09-17 Lexmark International, Inc. Ink jet swath prediction method
US6826436B1 (en) * 2001-04-25 2004-11-30 Autodesk, Inc. Method and apparatus for graphing motion of computer aided design (CAD) geometry
US7168038B2 (en) * 2001-08-01 2007-01-23 Microsoft Corporation System and method for scaling and repositioning drawings
US8824553B2 (en) * 2003-05-12 2014-09-02 Google Inc. Video compression method
JP2006013766A (en) * 2004-06-24 2006-01-12 Matsushita Electric Ind Co Ltd Method and device for recording, and method and device for reproducing moving image
US7953152B1 (en) 2004-06-28 2011-05-31 Google Inc. Video compression and encoding method
US7409099B1 (en) * 2004-08-10 2008-08-05 On2 Technologies, Inc. Method of improved image/video compression via data re-ordering
GB2431787B (en) * 2005-10-31 2009-07-01 Hewlett Packard Development Co A method of tracking an object in a video stream
JP4557959B2 (en) * 2006-12-14 2010-10-06 キヤノン株式会社 Electrophotographic image forming apparatus using toner and electrophotographic image forming method using toner
US8325796B2 (en) * 2008-09-11 2012-12-04 Google Inc. System and method for video coding using adaptive segmentation
US8326075B2 (en) * 2008-09-11 2012-12-04 Google Inc. System and method for video encoding using adaptive loop filter
US8311111B2 (en) 2008-09-11 2012-11-13 Google Inc. System and method for decoding using parallel processing
US8411970B2 (en) * 2010-03-16 2013-04-02 Pixia Corp. Method and system for determining statistical data for image pixels having a higher bit depth per band
US9106933B1 (en) 2010-05-18 2015-08-11 Google Inc. Apparatus and method for encoding video using different second-stage transform
CN102437999A (en) * 2010-09-29 2012-05-02 国际商业机器公司 Method and system for improving application sharing through dynamic partition
KR101712156B1 (en) * 2010-12-06 2017-03-06 에스케이 텔레콤주식회사 Method and Apparatus for Image Encoding/Decoding by Inter Prediction Using Arbitrary Shape Block
US9210442B2 (en) 2011-01-12 2015-12-08 Google Technology Holdings LLC Efficient transform unit representation
US9380319B2 (en) 2011-02-04 2016-06-28 Google Technology Holdings LLC Implicit transform unit representation
US8938001B1 (en) 2011-04-05 2015-01-20 Google Inc. Apparatus and method for coding using combinations
US8780996B2 (en) 2011-04-07 2014-07-15 Google, Inc. System and method for encoding and decoding video data
US8781004B1 (en) 2011-04-07 2014-07-15 Google Inc. System and method for encoding video using variable loop filter
US9154799B2 (en) 2011-04-07 2015-10-06 Google Inc. Encoding and decoding motion via image segmentation
US8780971B1 (en) 2011-04-07 2014-07-15 Google, Inc. System and method of encoding using selectable loop filters
US8885706B2 (en) 2011-09-16 2014-11-11 Google Inc. Apparatus and methodology for a video codec system with noise reduction capability
US9100657B1 (en) 2011-12-07 2015-08-04 Google Inc. Encoding time management in parallel real-time video encoding
US9262670B2 (en) 2012-02-10 2016-02-16 Google Inc. Adaptive region of interest
US9131073B1 (en) 2012-03-02 2015-09-08 Google Inc. Motion estimation aided noise reduction
US9344729B1 (en) 2012-07-11 2016-05-17 Google Inc. Selective prediction signal filtering
US9219915B1 (en) 2013-01-17 2015-12-22 Google Inc. Selection of transform size in video coding
US9544597B1 (en) 2013-02-11 2017-01-10 Google Inc. Hybrid transform in video encoding and decoding
US9967559B1 (en) 2013-02-11 2018-05-08 Google Llc Motion vector dependent spatial transformation in video coding
US9674530B1 (en) 2013-04-30 2017-06-06 Google Inc. Hybrid transforms in video coding
US11425395B2 (en) 2013-08-20 2022-08-23 Google Llc Encoding and decoding using tiling
US10127332B2 (en) * 2013-12-23 2018-11-13 Dassault Systemes Solidworks Corporation Automatic motion of a computer-aided design model
US20150262404A1 (en) * 2014-03-13 2015-09-17 Huawei Technologies Co., Ltd. Screen Content And Mixed Content Coding
US10102613B2 (en) 2014-09-25 2018-10-16 Google Llc Frequency-domain denoising
US9565451B1 (en) 2014-10-31 2017-02-07 Google Inc. Prediction dependent transform coding
US9769499B2 (en) 2015-08-11 2017-09-19 Google Inc. Super-transform video coding
US10277905B2 (en) 2015-09-14 2019-04-30 Google Llc Transform selection for non-baseband signal coding
US9807423B1 (en) 2015-11-24 2017-10-31 Google Inc. Hybrid transform scheme for video coding
US9794574B2 (en) 2016-01-11 2017-10-17 Google Inc. Adaptive tile data size coding for video and image compression
US10542258B2 (en) 2016-01-25 2020-01-21 Google Llc Tile copying for video compression
CN109754380B (en) * 2019-01-02 2021-02-02 京东方科技集团股份有限公司 Image processing method, image processing device and display device
US11122297B2 (en) 2019-05-03 2021-09-14 Google Llc Using border-aligned block functions for image compression

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5351083A (en) * 1991-10-17 1994-09-27 Sony Corporation Picture encoding and/or decoding system
EP0632662A1 (en) * 1993-06-29 1995-01-04 Koninklijke Philips Electronics N.V. Motion compensator for video coder/decoder
EP0639925A2 (en) * 1993-08-03 1995-02-22 Sony Corporation Motion vector detection

Family Cites Families (25)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US3873976A (en) * 1973-07-30 1975-03-25 Burroughs Corp Memory access system
US3943361A (en) * 1974-08-07 1976-03-09 Oregon Graduate Center For Study And Research Dimensional measurement of remote objects using projected light raster
FR2305802A1 (en) * 1975-03-26 1976-10-22 Pollard Jean PROCESS FOR MAKING A WRITING READABLE BY A SHAPE RECOGNITION MACHINE
US4463386A (en) * 1982-05-03 1984-07-31 International Business Machines Corporation Facsimile data reduction
US4608664A (en) * 1983-02-23 1986-08-26 International Business Machines Corporation Automatically balancing and vertically justifying a plurality of text/graphics-columns
US4626837A (en) * 1983-11-17 1986-12-02 Wyse Technology Display interface apparatus
US5046114A (en) * 1985-10-01 1991-09-03 The Palantir Corporation Method and structure for separating joined patterns for use in pattern and character recognition system
BE904101A (en) * 1986-01-24 1986-07-24 Bell Telephone Mfg Cy IMAGE PROCESSING SYSTEM AND PHASE LOCKING LOOP APPLIED THEREIN.
US5041921A (en) * 1987-01-06 1991-08-20 Duplitronics, Inc. System for recording custom albums from a library of pre-recorded items
US5621811A (en) * 1987-10-30 1997-04-15 Hewlett-Packard Co. Learning method and apparatus for detecting and controlling solder defects
US5617482A (en) * 1990-08-15 1997-04-01 Televerket Method of motion compensation and elastic deformation in picture sequences
DE69225920T2 (en) * 1991-03-06 1998-10-15 Mitsubishi Electric Corp Magnetic thin film memory device
US5418865A (en) * 1992-03-20 1995-05-23 Xerox Corporation Mark sensing on a form
KR0166717B1 (en) * 1992-06-18 1999-03-20 강진구 Encoding and decoding method and apparatus for variable dividing screen
US5471307A (en) * 1992-09-21 1995-11-28 Phase Shift Technology, Inc. Sheet flatness measurement system and method
JP3679426B2 (en) * 1993-03-15 2005-08-03 マサチューセッツ・インスティチュート・オブ・テクノロジー A system that encodes image data into multiple layers, each representing a coherent region of motion, and motion parameters associated with the layers.
US5384864A (en) * 1993-04-19 1995-01-24 Xerox Corporation Method and apparatus for automatic determination of text line, word and character cell spatial features
PT730465E (en) * 1993-09-24 2002-06-28 Univ Southern California UTILIZATION OF ANGIOTENSIN III AND ANALOGS OF THIS IN TISSUE SCISSORING
US5418659A (en) * 1993-12-23 1995-05-23 National Semiconductor Corporation Reliable time-domain demarcation of split formats in embedded-servo, zoned-data recording disk drives
US5611000A (en) * 1994-02-22 1997-03-11 Digital Equipment Corporation Spline-based image registration
US5650829A (en) * 1994-04-21 1997-07-22 Sanyo Electric Co., Ltd. Motion video coding systems with motion vector detection
KR0171143B1 (en) * 1995-03-20 1999-03-20 배순훈 Apparatus for composing triangle in the hexagonal grid
US5712657A (en) * 1995-03-28 1998-01-27 Cirrus Logic, Inc. Method and apparatus for adaptive dithering
US5706364A (en) * 1995-04-28 1998-01-06 Xerox Corporation Method of producing character templates using unsegmented samples
KR100235064B1 (en) * 1996-05-23 1999-12-15 전주범 Apparatus for coding objects using block based coding technique

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5351083A (en) * 1991-10-17 1994-09-27 Sony Corporation Picture encoding and/or decoding system
EP0632662A1 (en) * 1993-06-29 1995-01-04 Koninklijke Philips Electronics N.V. Motion compensator for video coder/decoder
EP0639925A2 (en) * 1993-08-03 1995-02-22 Sony Corporation Motion vector detection

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0809404A2 (en) * 1996-05-23 1997-11-26 Daewoo Electronics Co., Ltd Apparatus for coding an object region of a video signal by using a rearranged block-based technique
EP0809405A2 (en) * 1996-05-23 1997-11-26 Daewoo Electronics Co., Ltd Method and apparatus for determining an optimum grid for use in a block-based video signal coding system
EP0809405A3 (en) * 1996-05-23 1999-07-14 Daewoo Electronics Co., Ltd Method and apparatus for determining an optimum grid for use in a block-based video signal coding system
EP0809404A3 (en) * 1996-05-23 2003-11-19 Daewoo Electronics Corporation Apparatus for coding an object region of a video signal by using a rearranged block-based technique
WO2000025232A1 (en) * 1998-10-23 2000-05-04 Siemens Aktiengesellschaft Method and array for coding, decoding, and transmitting a digitized image
WO2002071759A1 (en) * 2001-03-07 2002-09-12 Nikon Corporation Image compression device, electronic camera, and image processing program
JP2002271790A (en) * 2001-03-07 2002-09-20 Nikon Corp Image compressor, electronic camera and image processing program
US7561748B2 (en) 2003-07-03 2009-07-14 Nikon Corporation Electronic camera

Also Published As

Publication number Publication date
JPH09510859A (en) 1997-10-28
CA2198463C (en) 2002-02-05
DE69626145T2 (en) 2003-12-11
TW320806B (en) 1997-11-21
DE69615805T2 (en) 2002-06-13
CA2198463A1 (en) 1997-01-30
US6115501A (en) 2000-09-05
EP1022910A2 (en) 2000-07-26
AU6319996A (en) 1997-02-10
EP1022910A3 (en) 2000-12-06
US5892849A (en) 1999-04-06
EP0781492A1 (en) 1997-07-02
JP3242409B2 (en) 2001-12-25
DE69615805D1 (en) 2001-11-15
US5903670A (en) 1999-05-11
US5917949A (en) 1999-06-29
EP0781492B1 (en) 2001-10-10
MY121607A (en) 2006-02-28
EP1022910B1 (en) 2003-02-05
DE69626145D1 (en) 2003-03-13

Similar Documents

Publication Publication Date Title
US6115501A (en) Grid moving method for minimizing image information of an object
US6738424B1 (en) Scene model generation from video for use in video processing
EP0284511B1 (en) Image information code processing system
CN1177259A (en) Encoding and decoding system of motion image containing arbitrary object
JP2787944B2 (en) Motion compensation motion vector search method
US5844631A (en) Motion estimation method and apparatus for calculating a motion vector
US6934332B1 (en) Motion estimation using predetermined pixel patterns and subpatterns
KR100248998B1 (en) Grid moving method of object image and apparatus using the same and compaction/motion estimation method using the same and apparatus thereof
US5859673A (en) Motion estimation method and apparatus for calculating a motion vector
US6813315B1 (en) Motion estimation using multiple search windows
KR100239291B1 (en) Grid moving method and apparatus using selective pixel investigation
CN1213231A (en) Method and apparatus for encoding motion vector
WO2001049028A1 (en) Scene model generation from video for use in video processing
KR100417132B1 (en) Coding method of object image and coding device
KR100420471B1 (en) Coding method and coding device of object image, particularly concerned with moving a macro block for dividing a vop in mpeg-4 to an information reduction position in accordance with shape information of an object image, to improve a coding efficiency
KR100424684B1 (en) Method and apparatus for encoding object image
US20070070059A1 (en) Refinement of block motion estimate using texture mapping
KR100424685B1 (en) Method and apparatus for encoding object image
US6891890B1 (en) Multi-phase motion estimation system and method
KR100424683B1 (en) Method and apparatus for encoding object image
KR100417128B1 (en) Encoding Method and Encoding Device of Object Image
KR100424686B1 (en) Method and apparatus for encoding object image
RU2212710C1 (en) Method for coding coordinates of video image moving on computer monitor screen
KR100417137B1 (en) Coding method of object image and coding device
JPH0779437A (en) Hierarchical motion compensation method for moving picture motion and vector interpolation predict method

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A1

Designated state(s): AL AM AT AU AZ BB BG BR BY CA CH CN CZ DE DK EE ES FI GB GE HU IS JP KE KG KZ LK LR LS LT LU LV MD MG MK MN MW MX NO NZ PL PT RO RU SD SE SG SI SK TJ TM TR TT UA UG UZ VN AM AZ BY KG KZ MD RU TJ TM

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): KE LS MW SD SZ UG AT BE CH DE DK ES FI FR GB GR IE IT LU MC NL PT SE BF BJ CF CG CI CM GA GN ML MR NE

WWE Wipo information: entry into national phase

Ref document number: 2198463

Country of ref document: CA

WWE Wipo information: entry into national phase

Ref document number: 1996922276

Country of ref document: EP

121 Ep: the epo has been informed by wipo that ep was designated in this application
WWP Wipo information: published in national office

Ref document number: 1996922276

Country of ref document: EP

REG Reference to national code

Ref country code: DE

Ref legal event code: 8642

WWG Wipo information: grant in national office

Ref document number: 1996922276

Country of ref document: EP