WO2001093188A9 - Procede de traitement de document, support sur lequel est enregistre un programme de traitement de document et processeur de document - Google Patents
Procede de traitement de document, support sur lequel est enregistre un programme de traitement de document et processeur de documentInfo
- Publication number
- WO2001093188A9 WO2001093188A9 PCT/JP2001/004673 JP0104673W WO0193188A9 WO 2001093188 A9 WO2001093188 A9 WO 2001093188A9 JP 0104673 W JP0104673 W JP 0104673W WO 0193188 A9 WO0193188 A9 WO 0193188A9
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- character
- input
- frame
- data
- frame data
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
- G09G3/00—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes
- G09G3/20—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes for presentation of an assembly of a number of characters, e.g. a page, by composing the assembly by combination of individual elements arranged in a matrix no fixed position being assigned to or needed to be assigned to the individual characters or partial characters
- G09G3/34—Control arrangements or circuits, of interest only in connection with visual indicators other than cathode-ray tubes for presentation of an assembly of a number of characters, e.g. a page, by composing the assembly by combination of individual elements arranged in a matrix no fixed position being assigned to or needed to be assigned to the individual characters or partial characters by control of light from an independent source
- G09G3/3406—Control of illumination source
- G09G3/3413—Details of control of colour illumination sources
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
- G06V30/148—Segmentation of character regions
- G06V30/155—Removing patterns interfering with the pattern to be recognised, such as ruled lines or underlines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
- G09G2310/00—Command of the display device
- G09G2310/02—Addressing, scanning or driving the display screen or processing steps related thereto
- G09G2310/0235—Field-sequential colour display
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09G—ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICES USING STATIC MEANS TO PRESENT VARIABLE INFORMATION
- G09G2320/00—Control of display operating conditions
- G09G2320/02—Improving the quality of display appearance
- G09G2320/0247—Flicker reduction other than flicker reduction circuits used for single beam cathode-ray tubes
Definitions
- Form processing method recording medium recording form processing program, form processing program, and form processing device
- the present invention relates to a form processing method, a recording medium recording a form processing program, a form processing program, and a form processing device.
- INDUSTRIAL APPLICABILITY The present invention is particularly applicable to the field of information equipment used in offices and the like, and a form processing method and a form that can easily recognize characters and symbols (hereinafter collectively referred to as letters) written on the form.
- the present invention relates to a recording medium on which a processing program is recorded and a form processing apparatus.
- the form refers to a two-dimensional shape or a three-dimensional shape such as a sheet formed of paper, a polymer material, or the like, having a surface on which an entry frame for entering characters is printed or described. Things shall be generically referred to.
- Fig. 37 shows the configuration of a conventional form processing device (see Japanese Patent Publication No. 63-187876).
- a text box is printed with a low density, and the text is written in the box.
- the photoelectric conversion means 2 photoelectrically converts the area of one line of the form 101 with the density of the character entry frame low and the density of the character high.
- the density of each pixel in the area for one row obtained as a result of the photoelectric conversion is stored in the storage means 103.
- the contents stored in the storage means 103 are sent to the character entry frame position detection means 104.
- the character entry frame position detecting means 104 counts the number of pixels having a predetermined density in the row and column directions, respectively.
- the character entry frame position detecting means 104 regards that there is a character entry frame at this position, and cuts out the character entry frame position information and extracts the character.
- the character cutout means 105 cuts out characters using the character entry frame position information and the storage content sent from the storage means 103.
- the entry frame indicates 1 or 0 indicating white or black in relation to the density and the binarization threshold. In some cases, depending on the binarization threshold, the entry frame disappears, or the entry frame becomes the same as the density of the entered characters. Or, in the past, a high-performance facsimile capable of accurately sending density information was required.
- the present invention is based on the Instead, a form processing method that outputs a set of small dots (called a dot texture) and erases the dot texture entry frame from a form with such an entry frame to retrieve only the entered characters, a form
- a processing program a recording medium on which a form processing program is recorded, and a form processing device.
- Another object of the present invention is to make a form using a black and white printer, since a form can be created using a line formed by a set of dots.
- Another object of the present invention is to make it possible to input a form using an inexpensive binary image reader, facsimile, scanner, or the like for reading. It is another object of the present invention to eliminate the need for line and character entry position marks, eliminate the need for special document output processing at the time of printing, and facilitate form creation.
- the form is usually printed with the heading of the item, such as above or to the left of the frame where the item is to be entered.
- headlines are separated from written characters by printing the headings in a dropout color like a frame, or by separating them with ambiguous information such as smaller than written characters.
- an object of the present invention is to print a heading character with dot texture in the same manner as a frame, so that the heading character is dropped in exactly the same manner as dropping a frame from an image.
- the present invention provides a method of separating a handwritten character pattern portion from an original image, then dropping the portion from the original image in reverse, and performing dilation processing on the remaining image, thereby enabling heading character recognition. Aim.
- the present invention expresses additional information (version, recognition instruction command, information used for authenticity determination, etc.) in the dot size, pitch, and number, thereby embedding information such as instructions to the dot texture. Aim.
- dots can be easily removed by shrinking or using conventional image processing technology such as a technology for measuring the number of connected black components (the number of connected black pixels surrounded by white pixels).
- image processing technology such as a technology for measuring the number of connected black components (the number of connected black pixels surrounded by white pixels).
- a form input step in which a form is input and a character frame data including a filled-in frame composed of a group of dots and filled-in characters is created;
- An entry frame detection step for detecting entry frame position data indicating a position where a character is to be entered, based on the character box data created in the form input step;
- Character detection means for detecting character data of the entered character based on the character frame data created by the form input means
- Entry frame detection means for detecting entry frame position data indicating a position where a character is to be entered, based on the character frame data created by the form input means;
- Recognizing means for recognizing each entered character based on each character pattern cut out by the associating means
- Headline character composed of a group of dots and entry composed of a group of dots
- a form processing method including: a form processing program for causing a computer to execute these steps; and a computer-readable recording medium storing the form processing program.
- An input box composed of a group of dots, in which information is embedded by changing the group of dots in a non-uniform manner, and inputting character frame data including the entered characters;
- FIG. 1 is a flowchart illustrating a form processing method according to the present invention.
- FIG. 2 is a configuration diagram of a form processing apparatus according to a first embodiment of the present invention.
- Figure 3 is an explanatory diagram related to form processing.
- Figure 4 is a detailed flowchart of the character detection process.
- FIG. 5 is an explanatory diagram of a character detection process.
- FIG. 6 is a detailed flowchart of the entry frame detection process.
- FIG. 7 is an explanatory diagram of the entry frame detection process.
- Figure 8 is a detailed flowchart of the association process.
- FIG. 9 is an explanatory diagram of the association processing.
- FIG. 10 is an explanatory diagram showing an example in which writing characters can be separated even if they overlap with each other in the present invention.
- FIG. 11 is a flowchart of a process according to the second embodiment for separating and recognizing a heading character.
- FIG. 12 is a configuration diagram of a form processing device according to a second embodiment of the present invention.
- Figure 13 is an illustration of an example of filling in a dot texture form.
- FIG. 14 is an explanatory diagram of extraction of a handwritten character pattern.
- Fig. 15 is an explanatory diagram of taking out the heading and the entry frame.
- FIG. 16 is a flowchart of processing in a case where the second embodiment of separation and recognition of a heading character is combined with the first embodiment.
- Figure 17 is a flowchart of dot removal using Fourier transform.
- Figure 18 is an explanatory diagram of the original image.
- FIG. 19 is an explanatory diagram when a discrete Fourier transform is performed on an original image.
- FIG. 20 is an explanatory diagram when high-frequency components are removed.
- FIG. 21 is an explanatory diagram when the inverse Fourier transform is performed.
- Fig. 2 2 is an explanatory diagram when the threshold is determined and binarized
- Figure 23 is a flowchart of dot removal using the Walsh-Rydamard transform.
- Figure 24 is an explanatory diagram of the original image.
- FIG. 25 is an explanatory diagram when the Walsh-Hadamard transform is performed on the original image.
- Figure 26 is an explanatory diagram when high-frequency components (and some low-frequency components) have been removed.
- FIG. 27 is an explanatory diagram when the inverse Walsh-Hadamard transform is executed.
- Figure 28 is an explanatory diagram when the threshold is determined and binarized.
- FIG. 29 is an explanatory diagram in the case where the low frequency is not cut off in the process of step S403.
- FIG. 31 is a configuration diagram of a form processing apparatus according to a third embodiment of the present invention.
- Figure 32 shows an example of a dot texture that forms this character frame.
- Figure 33 shows an example of adding information by changing the dot size.
- Figure 34 shows an example of adding information depending on the presence or absence of a dot.
- Figure 35 shows an example of embedding additional information by changing the dot pitch.
- FIG. 36 shows a flowchart of processing in a case where the third embodiment for embedding additional information and the first embodiment are combined.
- FIG. 37 shows a configuration diagram of a conventional form processing apparatus.
- FIG. 1 shows a flowchart of a form processing method according to the first embodiment of the present invention.
- FIG. 2 shows a configuration diagram of the form processing apparatus according to the first embodiment of the present invention.
- FIG. 3 shows an explanatory diagram relating to the form processing.
- This form processing apparatus includes a processing unit 11, an input unit 12, a storage unit 13, and an output unit 14.
- the processing unit 11 includes entry frame detecting means 4, character detecting means 5, associating means 6, and character recognizing means 7.
- the form input unit 12 includes a photoelectric conversion unit 2.
- the storage unit 13 includes a character frame data storage unit 3 and a result storage unit 8.
- the form input means 1 2 reads the form 1 to be recognized by the photoelectric conversion means 2 and obtains character frame data 11 1 including characters and entry boxes as electronic data (form input processing S 2) .
- the character frame data storage means 3 stores the character frame data 11 read by the photoelectric conversion means 2 (character frame data storage processing S 3).
- the character frame data 11 is detected by the frame detection means 4 as frame position data 13c indicating a position where a character is to be written (entry frame detection processing S4).
- the character data 12 b is detected by the character detection means 5 (character detection processing S 5).
- the character detection process S5 and the entry frame detection process S4 have been described as being processed in parallel, but they may be processed sequentially.
- the associating means 6 cuts out the character data 12b of the entered characters and the input frame position data 13c, and the character pattern entered in each of the input frames (association processing S6). As a result, the character pattern is recognized by the character recognition means 7 (character recognition processing S7), and the recognition result and the entry frame are associated with each other and stored in the result storage means 8 (result storage processing S8). At or after the result storage process S8, the output unit 14 may output or display a recognition result or data relating to the recognition result to another device.
- FIG. 4 shows a detailed flowchart of the character detection processing.
- FIG. 5 is an explanatory diagram of the character detection process.
- the contraction process is, for example, a process of contracting the periphery (edge) of a connected portion where black pixels are connected.
- a contracted image is obtained by extracting an edge of an input image and changing the edge to a white pixel.
- An edge is an extraction of a black pixel adjacent to a white pixel. Further, by extracting and removing double, triple, etc. edges, further contraction processing can be performed. Since the entry frame is made up of dots, this shrinking process can make the dots smaller, while the written characters are not significantly affected.
- minute points are removed from the contraction processing result 12a (S55).
- a connected component that is a portion where black pixels are connected is obtained, the number of black pixels of each connected component is obtained, and a connected component that is equal to or less than a predetermined threshold value or smaller than the threshold value is determined. Remove the minute. Next, by performing dilation on the remaining pattern in reverse
- character data 12b of the entered character can be obtained.
- expansion processing for example, an expanded image is obtained by extracting an edge of an input image and changing a white pixel adjacent to a black pixel at the edge to a black pixel.
- FIG. 6 shows a detailed flowchart of the entry frame detection processing.
- FIG. 7 shows an explanatory diagram of the entry frame detection process.
- the entry frame detection means 4 inputs the character frame data 11 (S4 1). Even if the vertical marginal distribution of the character frame data 11 is simply taken, there is an effect of the entered character, and the entry frame position cannot be accurately detected. Then, in order to remove the influence of the characters entered in the entry frame as much as possible, a process of removing the character pattern from the character frame data 11 is executed, and the entry frame data 13a is created (S43). 0 This process is performed by setting the pixel of the character frame data corresponding to the position of each black pixel of the character data 12b detected in the character detection process to a white pixel. Next, marginal distribution 13b is obtained by integrating or accumulating the entry frame data 13a in the vertical direction (S45). The entry frame position data 13c can be obtained from the peripheral distribution 13b (S47). At this time, for example, a delimiter of a fixed period correlated with the peak of the peripheral distribution 13b may be obtained.
- Another method to reduce the influence of characters entered in the entry frame is to first perform edge detection on the entire character frame data 11 and then take the vertical distribution around it to detect the entry frame position. There is also a way to do. By doing so, the process of removing the character pattern can be omitted.
- the dot detection generates a large number of edges (the number of black pixels that make up the edge) by edge detection, and on the other hand, the written characters generate only the edges corresponding to the outline, so the effect can be reduced.
- marginal distribution 13b is obtained by integrating or accumulating the entry frame data 13a in the row direction (S45).
- the entry frame position data 13c can be obtained from the peripheral distribution 13 (S47). At this time, for example, a fixed period break correlated with the peak of the peripheral distribution 13b may be obtained.
- FIG. 8 shows a detailed flowchart of the association processing.
- FIG. 9 shows an explanatory diagram of the association process.
- the associating means 6 includes the entry frame detecting process S 4
- the input frame position data 13c obtained in the above and the character data 12b obtained in the character detection processing S5 are input, and are associated with each other (S61).
- character patterns 14a to 14c are cut out for each character (S63).
- character patterns 14a to 14c can be cut out at the same cycle or length. Character recognition is performed based on the character patterns 14a to 14c for each character.
- the dot texture is applied to the entry frame as in the present invention, even if the writing character overlaps the entry frame, it can be easily separated by the above-described method.
- FIG. 10 is an explanatory diagram showing an example in which writing characters can be separated even if they overlap with each other in the present invention.
- the headline of the item is printed on the form above or to the left of the frame where the item is to be entered.
- the headlines were conventionally printed out with a dropout color like a frame and separated, or they were separated with ambiguous information such as smaller than the written characters.
- the headline characters by printing the headline characters with dot texture in the same manner as the frame, they can be dropped in exactly the same manner as dropping the frame from the image.
- the part After separating the handwritten character pattern part from the image, the part is dropped from the original image, and the remaining image is subjected to thickening processing (expansion processing), so that headline characters can be recognized.
- FIG. 11 shows a flowchart of a basic second embodiment of processing for separating and recognizing a heading character.
- FIG. 12 shows a configuration diagram of a form processing apparatus according to a second embodiment of the present invention.
- This form processing device has a processing unit 11, an input unit 12, A storage unit 13 and an output unit 14 are provided.
- the processing unit 11 includes an entry frame detection unit 4, a character detection unit 5, a correspondence unit 6, a character recognition unit 7, and a handwritten character removal unit 21.
- the form input unit 12 includes a photoelectric conversion unit 2.
- the storage unit 13 includes character frame data storage means 3 and result storage means 8.
- the configuration of each means is the same as that of the above-described embodiment, except for the contents described specifically, such as the heading character recognition means 22.
- Fig. 13 shows an explanatory diagram of an example of entry into a dot texture form.
- the processing unit 11 inputs the character frame data 2 11 1 as the original image (S 20 0 l) o
- the input unit executes the input processing as follows can do.
- the form input means 12 reads the form to be recognized by the photoelectric conversion means 2 and obtains character frame data 2 11 including characters and entry boxes as electronic data (FIG. 1, form input processing S 2).
- the character frame data storage means 3 stores the character frame data 211 read by the photoelectric conversion means 2 (see FIG. 1 character frame data storage processing S3).
- the processing unit 11 inputs the character frame data 2 11 from the character frame data storage unit 3 of the storage unit 13.
- the processing unit 11 performs a form input process S 2 and a character frame data storage process.
- the character frame data 2 1 1 may be directly input from the character frame data storage means 3 without executing S3.
- FIG. 14 shows an explanatory diagram for extracting a handwritten character pattern.
- the character detection means 5 performs a character detection process on the handwritten character pattern (S203). More specifically, the character detecting means 5 includes a detailed flowchart of the character detecting process shown in FIG. 4 described above, an explanatory diagram of the character detecting process shown in FIG. 5 described above, and descriptions thereof. Perform the character detection process as described in the section. As a result, character data 211 of the character entered by the character detecting means 5 can be obtained.
- the handwritten pattern removing means 21 of the processing unit 11 removes the character pattern 2 1 2 detected in the character detection processing S 203 from the character frame data 2 1 1 based on the input character frame data 2 1 1.
- entry frame data including a heading (S205).
- Figure 15 shows an explanatory diagram for taking out the headline and the entry frame.
- the handwriting pattern removing means 21 applies dot texture to the obtained entry frame data.
- the expansion processing of each dot is performed to the extent that is connected (S207). As a result, the heading and the entry frame as shown in the figure are taken out.
- the character recognition means 7 is retrieved heading (or headline entry frame) with respect to, that by the the performed (S 2 0 9) 0 character recognition means 7 minute point recognition headline character noise If it is weak, the minute point removal (S55) described with reference to FIG. 4 may be performed before or after the expansion processing in the handwritten pattern removing means.
- FIG. 16 shows a flowchart of a process in a case where the second embodiment for separating and recognizing a heading character is combined with the first embodiment.
- a form input process S1 and a character frame data storage process S2 are executed.
- the entry frame detection processing S4, character detection processing S5, and association processing S6 are as described above.
- the handwritten pattern removing means 21 reads out the character frame data accumulated by the character frame data accumulation processing S3 from the character frame data accumulation means 3 and inputs it as an original image (S21). 1).
- the handwriting pattern removing unit 21 removes the handwritten character pattern as described above using the character data obtained by the character detecting process S5 (character detecting unit 5) (S215) .
- the handwritten pattern removing means 21 performs dilation processing on the entry frame data including the obtained heading (S2177).
- the handwritten pattern removing means 21 executes the processing for separating the heading characters in this manner. Further, in the character recognition process S7, in addition to the handwritten characters obtained in the associating process S6, the character recognition process is also performed for the headline character separated from step S2117. Further, in the character recognition processing means 7, the recognized handwritten characters and heading characters are stored in the result storage means 8. If the character recognition means 7 is vulnerable to noise due to minute points, the minute points removal (S55) described in FIG. 4 may be performed before or after the expansion processing in the handwritten pattern removing means.
- dot textures can be It is also possible to easily separate the editing symbols and the like for them and apply the corrections that the editing symbols mean.
- a dot texture is composed of higher frequency components than a handwritten pattern. Therefore, Fourier transform is applied to the input character frame data (image) to convert it to frequency components, where high-frequency components are removed, and inverse transform and binarization are performed to remove dots. There is a way to do it.
- FIG. 17 shows a flowchart of dot removal using the Fourier transform.
- the character detecting means 5 reads the character frame data which is the original image data stored in the character frame data storing means 3 (S301).
- the form input processing S2 and the character frame data storage processing S3 store the character frame data in the character frame data storage means 3, and the character detection means 5 inputs the character frame data. You may do it.
- Figure 18 shows the illustration of the original image.
- the character detecting means 5 executes a discrete Fourier transform on the read original image (S302).
- the original image data includes, for example, data representing grayscale or black and white when an image is scanned for a plurality of scans.
- the processing here is to perform a discrete Fourier transform of the change in grayscale or black and white.
- Figure 19 shows an illustration when the discrete Fourier transform is performed on the original image. In this figure, the center is the origin, the vertical axis is the vertical frequency component, and the horizontal axis is the horizontal frequency component.
- FIG. 20 shows an explanatory diagram for removing high-frequency components.
- the center is the origin
- the vertical axis is the vertical frequency component
- the horizontal axis is the horizontal frequency component.
- the character detecting means 5 executes an inverse discrete Fourier transform (S304).
- FIG. 21 shows an explanatory diagram when the inverse discrete Fourier transform is executed.
- the character detection means 5 determines a threshold value for the blurred image obtained in step S304 and performs a binarization process. Perform (S305).
- Fig. 22 shows an explanatory diagram when the threshold is determined and binarized.
- the Orche-Hadamard transform decomposes into square-wave frequency components.
- the original image is a binary image, this can represent the original image more faithfully even in the finite frequency range.
- a method of transforming into a rectangular wave frequency component by the Walsh-Hadamard transform, removing the high-frequency component therefrom, inversely transforming it, and binarizing it to remove the dot is also used. Conceivable.
- Figure 23 shows a flowchart of dot removal using the Walsh-Hadamard transform.
- the character detecting means 5 reads the character frame data which is the original image data stored in the character frame data storage means 3 (S401).
- the form input processing S2 and the character frame data storage processing S3 store the character frame data in the character frame data storage means 3, and the character detection means 5 inputs the character frame data. You may do it.
- Figure 24 shows an explanatory diagram of the original image.
- the character detecting means 5 executes Walsh-Hadamard transformation on the read original image (S402).
- the original image data includes, for example, data representing black and white when an image is scanned for a plurality of scans, and the processing here is to perform a Walsh-Hadamard transformation of the change in black and white.
- FIG. 25 shows an explanatory diagram when the Walsh-Hadamard transform is performed on the original image. In this figure, the center is the origin, the vertical axis is the vertical frequency component, and the horizontal axis is the horizontal frequency component.
- the character detecting means 5 executes the removal of the high frequency component (S403). This process blurs the original image. At this time, in some cases, low-frequency components may be removed at least in part.
- Figure 26 shows an explanatory diagram when high-frequency components (and some low-frequency components) have been removed. In this figure, the center is the origin, the vertical axis represents the vertical frequency component, and the horizontal axis represents the horizontal frequency component.
- the character detecting means 5 executes the inverse Walsh-Hadamard transform (S404).
- FIG. 27 is an explanatory diagram when the inverse Walsh-Hadamard transform is executed.
- the character detection means 5 determines a threshold value for the blurred image obtained in step S404 and performs a binarization process (S405).
- Figure 28 shows an explanatory diagram when the threshold is determined and binarized.
- FIG. 29 is an explanatory diagram in the case where the low frequency is not cut off in the process of step S403. That is, it is a binarized character pattern obtained through steps S404 and S405.
- the Fourier transform and Walsh-Hadamard transform have in common.
- the processing time can be made shorter by using dedicated hardware that performs Fourier transform and Walsh-Hadamard transform than by contraction, minute point removal, and expansion.
- FIG. 30 shows a flowchart for embedding information in a dot texture.
- FIG. 31 shows a configuration diagram of a form processing apparatus according to a third embodiment of the present invention.
- This form processing device includes a processing unit 11, an input unit 12, a storage unit 13, and an output unit 1. 4 is provided.
- the processing unit 11 includes an entry frame detection unit 4, a character detection unit 5, a correspondence unit 6, a character recognition unit 7, and an additional information determination unit 23.
- the form input section 12 includes photoelectric conversion means 2.
- the storage unit 13 includes a character frame data storage unit 3 and a result storage unit.
- the processing section 11 inputs character frame data 2 11 as an original image (S501).
- the input processing can be executed in the same manner as in the above-described embodiment (see step S201).
- the character detection means 5 performs a character detection process on the handwritten character pattern (S503). Specifically, as in step S203 described above, character data 211 of the character entered by the character detecting means 5 can be obtained.
- the handwriting pattern removing means 21 of the processing unit 11 removes the character pattern 2 12 detected by the character detection processing S 203 from the character frame data based on the input character frame data, thereby obtaining a heading. Is obtained (S505).
- the additional information determination means 23 determines the additional information embedded in the dot texture such as the obtained entry frame data in a predetermined format (S507).
- the additional information determination unit 23 stores the extracted additional information in the storage unit 13 (S509).
- Fig. 32 shows an example of the dot texture that forms this character frame.
- Each means is the same as the above-described embodiment, except for the contents described in particular such as the additional information determination means 23.
- Examples of the method of inserting the additional information include, for example, character type information to be entered such as numbers and kanji, character data types to be entered such as addresses, names and years, or character entry frames. It is appropriate information such as information on handwritten characters and headings. Information can be added to the character entry frame by changing the size, position, etc. of some of the dots in the dot texture shown in the figure. In addition, by embedding this information in a plurality of positions of the character frame, information can be stably taken out even when a part is dirty. Specific examples are shown below.
- FIG. 33 shows an example in which information is added according to a change in dot size.
- large Information can be embedded in a part of the character frame by setting the critical dot to 1 and the small dot to 0 (or vice versa).
- information [1 0 0 1 1 0 1 0] is embedded in all rows.
- FIG. 34 shows an example in which information is added depending on the presence or absence of a dot. For example, it is possible to embed information in a part of the character frame by setting 1 when there is a dot and 0 when there is no dot. In the example shown in the figure, information [1 0 0 0 1 1 1] is embedded in the line indicated by the arrow.
- FIG. 35 shows an example in which additional information is embedded by a change in dot pitch.
- information can be embedded in a part of the character frame by setting the dot pitch to 1 if it is wide and 0 if it is narrow.
- information [0 1 1 0 1 1 1 0 0 1] is embedded in all rows.
- the additional information processing means 23 can read the additional information by these predetermined methods based on the dot texture of the entry frame and the like extracted by the entry frame detection means 4 as described above.
- FIG. 36 shows a flowchart of processing in a case where the third embodiment for embedding additional information and the first embodiment are combined.
- a form input process S1 and a character frame data storage process S2 are executed.
- the entry frame detection processing S4, character detection processing S5, and association processing S6 are as described above.
- the handwritten pattern removing means 21 reads out the character frame data accumulated by the character frame data accumulation processing S3 from the character frame data accumulation means 3 and inputs it as an original image (S31). 1).
- the handwritten pattern removing unit 21 removes the handwritten character pattern as described above using the character data obtained by the character detecting process S5 (character detecting unit 5) (S315) .
- the additional information judging means 23 judges the additional information embedded in the dot texture such as the obtained entry frame data (S3 17). Further, the additional information determination means 23 stores the determined additional information in the result storage means 8.
- the dot frame is not a solid black unless the outline font, JP2001 / 004673
- a form processing method includes: a computer-readable recording medium storing a form processing program; a program product including the form processing program and capable of being loaded into an internal memory of a computer; It can be provided by a program product or the like stored on a medium.
- INDUSTRIAL APPLICABILITY According to the present invention, as described above, the entry frame is output with dots instead of color ink or gray color, and the entry frame of dots is erased from the form having such an entry frame and entered. It is possible to provide a form processing method capable of extracting only written characters, a recording medium recording a form processing program, and a form processing apparatus.
- a form can be created using a line formed by a set of dots, it is possible to create a form with a black and white printer. Further, according to the present invention, it is possible to input a form using a general image reader, FAX, scanner, or the like for reading. Furthermore, according to the present invention, it is possible to eliminate the need for line and character entry position marks, eliminate the need for special document output processing at the time of printing, and facilitate form creation.
- the present invention by printing the headline characters with dot texture in the same manner as the frame, it is possible to drop the frame in exactly the same manner as dropping the frame from the image. Also, according to the present invention, after separating a handwritten character pattern portion from the original image, the portion is dropped from the original image in reverse, and the remaining image is subjected to a thickening process, thereby enabling recognition of a found character. be able to. Furthermore, according to the present invention, by embedding information such as instructions to dot textures by expressing additional information (eg, information used for authenticity judgment, purge instructions, recognition instructions, etc.) in the size, pitch, and number of dots. Can be.
- additional information eg, information used for authenticity judgment, purge instructions, recognition instructions, etc.
Description
Claims
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP01936835A EP1310904A4 (en) | 2000-06-02 | 2001-06-01 | DOCUMENT PROCESSING METHOD, RECORDING MEDIA RECORDING A DOCUMENT PROCESSING PROGRAM AND DOCUMENT PROCESSOR |
US10/276,329 US7221795B2 (en) | 2000-06-02 | 2001-06-01 | Document processing method, recording medium having recorded thereon document processing program, document processing program, document processing apparatus, and character-input document |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2000-166025 | 2000-06-02 | ||
JP2000166025 | 2000-06-02 |
Publications (2)
Publication Number | Publication Date |
---|---|
WO2001093188A1 WO2001093188A1 (fr) | 2001-12-06 |
WO2001093188A9 true WO2001093188A9 (fr) | 2004-11-04 |
Family
ID=18669404
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2001/004673 WO2001093188A1 (fr) | 2000-06-02 | 2001-06-01 | Procede de traitement de document, support sur lequel est enregistre un programme de traitement de document et processeur de document |
Country Status (3)
Country | Link |
---|---|
US (1) | US7221795B2 (ja) |
EP (1) | EP1310904A4 (ja) |
WO (1) | WO2001093188A1 (ja) |
Families Citing this family (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP4704601B2 (ja) * | 2000-11-01 | 2011-06-15 | 富士通株式会社 | 文字認識方法,プログラム及び記録媒体 |
JP4777024B2 (ja) * | 2005-09-06 | 2011-09-21 | キヤノン株式会社 | 画像処理装置および画像処理装置の制御方法 |
CN101354746B (zh) * | 2007-07-23 | 2011-08-31 | 夏普株式会社 | 文字图像抽出装置及文字图像抽出方法 |
CN101562691B (zh) * | 2008-04-17 | 2010-12-08 | 鸿富锦精密工业(深圳)有限公司 | 图像处理装置及方法 |
CN102033928B (zh) * | 2010-12-15 | 2013-01-23 | 北京壹人壹本信息科技有限公司 | 手写原笔迹记录及查询系统、电子装置 |
JP6303814B2 (ja) * | 2014-05-28 | 2018-04-04 | 沖電気工業株式会社 | 認識装置、認識方法、プログラムおよび認識システム |
JP6903966B2 (ja) * | 2017-03-16 | 2021-07-14 | 富士フイルムビジネスイノベーション株式会社 | 情報処理装置、情報処理システム及びプログラム |
Family Cites Families (24)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS5654576A (en) * | 1979-10-12 | 1981-05-14 | Toshiba Corp | Optical character reader |
JPS56101273A (en) | 1980-01-18 | 1981-08-13 | Mitsubishi Electric Corp | Character disassembling system |
US4817179A (en) * | 1986-12-29 | 1989-03-28 | Scan-Optics, Inc. | Digital image enhancement methods and apparatus |
DE3880847T2 (de) * | 1987-01-20 | 1993-11-18 | British Tech Group | Verfahren und Vorrichtung zur Informationsergreifung beim Zeichnen oder Schreiben. |
US4883291A (en) * | 1988-05-11 | 1989-11-28 | Telesis Controls Corporation | Dot matrix formed security fonts |
JPH0632078B2 (ja) * | 1988-12-09 | 1994-04-27 | 工業技術院長 | 画像信号の処理方法 |
US5448692A (en) * | 1991-03-27 | 1995-09-05 | Ricoh Company, Ltd. | Digital image processing device involving processing of areas of image, based on respective contour line traces |
JPH06506080A (ja) * | 1991-04-03 | 1994-07-07 | ヒューレット・パッカード・カンパニー | 位置検知装置 |
JPH0528313A (ja) | 1991-07-18 | 1993-02-05 | Hitachi Electron Eng Co Ltd | 光学情報読取装置 |
JPH05113879A (ja) * | 1991-10-23 | 1993-05-07 | T A S Tsusho Kk | プログラムの作成及びデータの処理方法 |
JP2789971B2 (ja) * | 1992-10-27 | 1998-08-27 | 富士ゼロックス株式会社 | 表認識装置 |
JPH06231305A (ja) * | 1993-02-05 | 1994-08-19 | Fujitsu Ltd | 文字認識方法、および該方法に使用する帳票 |
JP3655651B2 (ja) * | 1994-09-02 | 2005-06-02 | テキサス インスツルメンツ インコーポレイテツド | データ処理装置 |
JPH08190606A (ja) * | 1995-01-10 | 1996-07-23 | Toshiba Corp | 光学的文字読取装置 |
JPH08194777A (ja) | 1995-01-13 | 1996-07-30 | Nippon Telegr & Teleph Corp <Ntt> | 手書き文字入力処理方法および手書き文字認識処理方法 |
JPH08329187A (ja) * | 1995-06-06 | 1996-12-13 | Oki Electric Ind Co Ltd | 文書読取装置 |
FR2737930B1 (fr) * | 1995-08-18 | 1997-10-31 | Itesoft | Procede et systeme de reconnaissance de caracteres manuscrits |
JPH09185683A (ja) * | 1995-12-28 | 1997-07-15 | Advantest Corp | 文字検査装置の画像処理方法とその装置 |
US6104833A (en) * | 1996-01-09 | 2000-08-15 | Fujitsu Limited | Pattern recognizing apparatus and method |
JP3615333B2 (ja) * | 1996-12-05 | 2005-02-02 | 株式会社リコー | 罫線消去装置 |
JPH10171996A (ja) * | 1996-12-09 | 1998-06-26 | Advantest Corp | 画像内粒状点検出方法 |
US6327387B1 (en) * | 1996-12-27 | 2001-12-04 | Fujitsu Limited | Apparatus and method for extracting management information from image |
JPH11143986A (ja) * | 1997-10-17 | 1999-05-28 | Internatl Business Mach Corp <Ibm> | ビットマップイメージの処理方法及び処理装置、ビットマップイメージの処理を行うイメージ処理プログラムを格納した記憶媒体 |
JP3968614B2 (ja) * | 1998-08-06 | 2007-08-29 | カシオ計算機株式会社 | 情報転送装置及び情報転送プログラムを記録した記録媒体 |
-
2001
- 2001-06-01 US US10/276,329 patent/US7221795B2/en not_active Expired - Fee Related
- 2001-06-01 EP EP01936835A patent/EP1310904A4/en not_active Withdrawn
- 2001-06-01 WO PCT/JP2001/004673 patent/WO2001093188A1/ja not_active Application Discontinuation
Also Published As
Publication number | Publication date |
---|---|
EP1310904A1 (en) | 2003-05-14 |
US20030113019A1 (en) | 2003-06-19 |
EP1310904A9 (en) | 2004-01-28 |
WO2001093188A1 (fr) | 2001-12-06 |
EP1310904A4 (en) | 2007-03-28 |
US7221795B2 (en) | 2007-05-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
KR101016712B1 (ko) | 워터마크 정보검출 방법 | |
JP4189506B2 (ja) | 画像処理のための装置、方法及び記録媒体 | |
EP1310912A2 (en) | Image processing method, apparatus and system | |
US8611658B2 (en) | Image processing apparatus and image processing method | |
JP2000194850A (ja) | ユ―ザ囲み領域の抽出装置及び抽出方法 | |
JPS63158678A (ja) | 単語間スペ−ス検出方法 | |
US7965293B2 (en) | Image processing device, image processing method, and image processing program for reconstructing data | |
US8229214B2 (en) | Image processing apparatus and image processing method | |
EP1310940A1 (en) | Color display device and method | |
WO2001093188A9 (fr) | Procede de traitement de document, support sur lequel est enregistre un programme de traitement de document et processeur de document | |
US8477988B2 (en) | Image processing apparatus, image forming apparatus and computer readable medium | |
US7142733B1 (en) | Document processing method, recording medium recording document processing program and document processing device | |
US6983071B2 (en) | Character segmentation device, character segmentation method used thereby, and program therefor | |
JP3728209B2 (ja) | 画像処理方法及び装置及びコンピュータプログラム及び記憶媒体 | |
WO2000060531A2 (en) | Extracting user data from a scanned image of a pre-printed form | |
JP4804433B2 (ja) | 画像処理装置、画像処理方法、及び、画像処理プログラム | |
JPS6325391B2 (ja) | ||
CN100511267C (zh) | 图文影像处理装置及其影像处理方法 | |
JP4552757B2 (ja) | 画像処理装置、画像処理方法、および画像処理プログラム | |
JP3720748B2 (ja) | 画像処理装置及びその制御方法並びにコンピュータプログラム及び記録媒体 | |
JP2708604B2 (ja) | 文字認識方法 | |
JP3756660B2 (ja) | 画像認識方法、装置および記録媒体 | |
JPS6254380A (ja) | 文字認識装置 | |
JP3566738B2 (ja) | 網掛け領域処理方法および網掛け領域処理装置 | |
JP4738645B2 (ja) | 網掛け領域検出装置、網掛け領域検出方法、プログラムおよび記憶媒体 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AK | Designated states |
Kind code of ref document: A1 Designated state(s): JP US |
|
AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): DE FR GB IT |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
DFPE | Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101) | ||
WWE | Wipo information: entry into national phase |
Ref document number: 2001936835 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 10276329 Country of ref document: US |
|
WWP | Wipo information: published in national office |
Ref document number: 2001936835 Country of ref document: EP |
|
COP | Corrected version of pamphlet |
Free format text: PAGE 2/37, DRAWING, REPLACED BY CORRECT PAGE 2/37 |
|
WWW | Wipo information: withdrawn in national office |
Ref document number: 2001936835 Country of ref document: EP |