CA2157711C - Anti-counterfeit pattern detector and method - Google Patents
Anti-counterfeit pattern detector and methodInfo
- Publication number
- CA2157711C CA2157711C CA002157711A CA2157711A CA2157711C CA 2157711 C CA2157711 C CA 2157711C CA 002157711 A CA002157711 A CA 002157711A CA 2157711 A CA2157711 A CA 2157711A CA 2157711 C CA2157711 C CA 2157711C
- Authority
- CA
- Canada
- Prior art keywords
- orientation
- phi
- templates
- theta
- image
- Prior art date
- Legal status (The legal status 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 status listed.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 20
- 238000001514 detection method Methods 0.000 abstract description 7
- 238000013527 convolutional neural network Methods 0.000 abstract description 3
- 238000005070 sampling Methods 0.000 description 4
- 238000013519 translation Methods 0.000 description 4
- 230000014616 translation Effects 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07D—HANDLING OF COINS OR VALUABLE PAPERS, e.g. TESTING, SORTING BY DENOMINATIONS, COUNTING, DISPENSING, CHANGING OR DEPOSITING
- G07D7/00—Testing specially adapted to determine the identity or genuineness of valuable papers or for segregating those which are unacceptable, e.g. banknotes that are alien to a currency
- G07D7/06—Testing specially adapted to determine the identity or genuineness of valuable papers or for segregating those which are unacceptable, e.g. banknotes that are alien to a currency using wave or particle radiation
- G07D7/12—Visible light, infrared or ultraviolet radiation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/24—Aligning, centring, orientation detection or correction of the image
- G06V10/245—Aligning, centring, orientation detection or correction of the image by locating a pattern; Special marks for positioning
Abstract
An anti-counterfeit detector and method for identifying whether a platen image portion to be photocopied contains one or several pre-selected monetary note patterns. The detection is performed in a rotation and shift invariant manner. Specifically, the pattern can be of any orientation and at any location of the image. Moreover, it can be embedded in any complicated image background.
Description
ANTI-COUNTERFEIT PATTERN DETECTOR AND METHOD
BACKGROUND OF THE INVENTION
The present invention relates to an anti-counterfeit detector and method for identifying whether an image to be photocopied contains one or several pre-selected monetary note patterns.
Preventing colour copiers from being misused for counterfeiting has currently drawn more and more attention. In determining whether a colour copier is being used for counterfeiting, a detector compares a known currency image with an image begin copied. A problem arises in that it is difficult to detect the patterns in a rotation and shift invariant manner.
Specifically, the pattern could be of any orientation and at any location on the image. The orientation and the location of the note can be relatively simple to obtain in the case of a single note with a plain background. However, it is difficult to obtain orientation and location if multiple notes are involved and/or the notes are embedded in some complicated image background.
SUMMARY OF THE INVENTION
It is therefore an object of an aspect of the present invention to provide an anti-counterfeit pattern detector that detects patterns in a rotation and shift invariant manner.
It is a further object of an aspect of the present invention to provide greater accuracy and lower analysis time for the currency detection and orientation process.
Further aspects of the present invention are as follows:
An anti-counterfeit detector for detecting whether an image to be photocopies is a currency note, the detector comprising:
storage means for storing a plurality of templates, each of said plurality of templates comprising at least one predetermined anchor point;
r - la -examining means for examining a portion of said image to be photocopied and for determining whether said portion contains a predetermined pixel intensity arrangement;
orientation determining means for determining an orientation of said predetermined pixel intensity arrangement;
positioning means for positioning said template in alignment with said predetermined pixel intensity arrangement in accordance with said orientation determined by said orientation determining means; and matching means for comparing said plurality of templates to said predetermined pixel intensity arrangement;
wherein said predetermined pixel intensity arrangement is an edge and said examining means comprises means for calculating edge strength within said portion in accordance with the following equations:
f (x. Y~ - ~ d~ ~ 3 (x, Y~ + ~ dH ~ 3 (x, Y~ , and w = Ef (x, y) where d~ and dH are differences in intensity of neighbouring pixels in vertical and horizontal directions respectively, W is edge strength and summation is over said portion.
An anti-counterfeit detector for detecting whether an image to be photocopied is a currency note, the detector comprising:
a memory for storing a plurality of templates, each of said plurality of templates comprising at least one predetermined anchor point;
a scanner controller that examines a portion of said image to be photocopied and that determines whether said portion contains a predetermined pixel intensity arrangement;
an orientation detector operatively coupled to said scanner controller that detects an orientation of said portion;
a - lb -a positioning controller for aligning orientation of said template to said orientation of said portion; and a matching controller operatively coupled to said memory, said scanner controller, positioning controller and said orientation detector, said matching controller comparing said plurality of templates to said portion in accordance with said orientation;
wherein said predetermined pixel intensity arrangement is an edge and said examining means comprises means for calculating edge strength within said portion in accordance with the following equations:
f (x, Y) - ~ d~ ~ 3 (x, y) + ~ dH ~ 3 (x, y) , and w = Ef (x, y) where d~ and dH are differences in intensity of neighbouring pixels in vertical and horizontal directions respectively, W is edge strength and summation is over said portion.
An anti-counterfeit method for detecting whether an image to be photocopied is a currency note, the method comprising the steps of:
storing a plurality of templates, each of said plurality of templates comprising at least one predetermined anchor point;
examining a portion of said image to be photocopied and determining whether said portion contains a predetermined pixel intensity arrangement;
determining an orientation of said predetermined pixel intensity arrangement;
positioning said templates in alignment with said predetermined pixel intensity arrangement in accordance with said orientation determined by said orientation determination means; and comparing said plurality of templates to said predetermined pixel intensity arrangement;
A
wherein said predetermined pixel intensity arrangement comprises an edge and the step of examining comprises determining edge strength within said portion in accordance with the following equations:
f (x, Y) - ~ dV ~ 3 (x, y) + ~ dH ~ 3 (x, y) , and W = Ef (x, y) where d~ and dH are differences in intensity of neighbouring pixels in vertical and horizontal directions respectively, W is edge strength and summation is over said portion.
BRIEF DESCRIPTION OF THE DRAWINGS
These and other aspects of the present invention will become apparent from the following detailed description of the preferred embodiments when taking in conjunction with the accompanying drawings, in which:
A
2157'11 _2_ Figures lA-D illustrate matching a template to a detected sample;
Figure 2 is a block diagram of the system of the invention;
Figure 3 is a flowchart illustrating the operation in the invention; and Figure 4 is a flowchart illustrating orientation estimation according to the invention.
DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS
The following detailed description of preferred embodiments is applicable to numerous detection and imaging systems as would be contemplated by those of ordinary skill.
The invention will be described with general reference to Fig. 3 .
In order to detect whether a color copier is being used for counterfeiting, the detector 1 is first trained off-line with example notes.
A template is created by an initial sampling of currencies (hereinafter template) suspected of being photocopied and counterfeited.
Conventional sampling is preferably performed in a low resolution.
The training includes sampling the templates and selecting one or several anchor points. Only those pixels on relatively straight edges, preferably long straight edges, are qualified to be anchor points. Anchor points are used to determine the edge orientation of the unknown document or documents and orient the template accordingly (discussed below). There is no set limit as to the number of possible anchor points in each template. The greater the number of preselected anchor points, the higher the detection accuracy, but the more time of analysis.
One-dollar bills may be sampled at 16 dpi and 32 dpi, respectively. That resolution is sufficient to make a positive determination of the presence of currency when the template is compared with the image being photocopied. The template is stored in a computer memory 26 (Fig. 2) for future image comparison with the unknown document.
The detecting of currency notes placed on a platen will now be described. A digital color copier contains functionally a scanning part 20 and a printing part 28 as indicated in Fig.. 2. The currency detector 1 is placed in parallel to the normal video pass 3o. A data processor (CPU) 22 performs the function of the detector 1.
21~~711 When an image of the platen (platen image) containing the unknown document is scanned by the scanning part of the copier, the signal is also sent to the detector 1. The image is sampled using conventional means and at the resolution of the template (Fig. 3, step 51), and stored into the image buffer 21.
The information of a scanned color image is typically organized into three or four channels. The most commonly used sets of channels, or color spaces, are RGB and CIELAB. In RGB, three channels carry red (R), green (G), and blue (B) signals, respectively, while in CIELAB, L* channel represents luminance information, and a* and b* channels represent chromanance information. If RGB space is used, all three channels will be sampled at the same resolution of 16 dpi or 32 dpi. If L*a*b* space is used, a* and b* channels are sampled at a resolution that is half of that of the L*
channel. Specifically, L* is sampled at l6dpi or 32 dpi, and a* and b* are sampled at 8dpi or l6dpi.
After sampling, the color image of the platen is smoothed (Fig.
3, step S2). The process of smoothing entails the averaging of the value of pixels within a given area and reassigning the averaged value to the center pixel. For example, a block area of 3 x 3 pixels may contain pixels of different value. An average value is obtained for the nine pixels, and is assigned to the center pixel. The reassigned value produces a new version of the platen image providing less prominent fine texture patterns that could otherwise confuse edge detection and orientation estimation (see below).
The smoothed platen image is examined block by block, with a typical block size of 8 x 8 pixels. The blocks can be overlapping. Each block is examined by the data processor 22 (Fig. 2 to see if it possibly contains a pixel intensity orientation that corresponds to a preselected anchor point on the template. The "quiet blocks" containing little pixel variation, can be initially discarded as an edge is not present within the block (Fig. 3 , step 53). For remaining blocks, the orientation of the edges contained in the block is estimated. The edge strength and orientation estimation is only performed on one of the color channels of the smoothed scanned image.
In RGB space it is on G channel, and in CIELAB space it is on L* channel.
The currency images often contain many fine textures that could be mistaken as edges. It is therefore very important to reduce the influence of texture during the edge strength and orientation estimation.
The edge strength at a pixel (x,y) is quantified by the following formula:
f(x,Y) = Idvl3(x,Y) + IdHl3(x,Y) ...(1) ._ wl~'~711 where d~ and dH are differences in intensity of neighboring pixels of the smoothed image in the vertical and horizontal directions, respectively. The formula emphasizes strong differences, which typically characterize edges, more than the weak differences, which usually represent the texture.
The edge strength of the block is determined by the following equation:
W = Ff(x,y), ...(2) where the summation is over the 8x8 pixel block.
The determined edge strength W is measured against a pre-chosen threshold. (Fig. 3 step 54). A block with weighted edge strength in excess of the threshold contains a major edge. If the edge strength does not exceed the pre-chosen threshold, the block is discarded, and a subsequent block is examined (Fig.3,step S5).
If it is determined that a block contains an edge, before it can be compared (matched) with the stored templates, the orientation of the edge must be determined (Fig.3 , step S6).
Estimation of edge orientation (Fig.3, step S6) will now be described with reference to Fig. 4 , For the detector to properly determine counterfeiting, orientation estimation must be reliable and relatively accurate. In the present invention, a method based on second order moments is utilized, which is accurate and robust to noise.
First, a candidate for the anchor point is found by calculating xo = [Ef(x,y)x]/W, and Yo = IEf(x.Y)Y]~W, ...(3) where the summations are over the 8x8 block (Fig. 4 , step 5100). The orientation of the edge is then evaluated using the following equation (Fig. 4, step 5103):
~=8+0.5n, ifs>0.5;
8, otherwise, ...(4) where A = 0.5tan-~[2Mxy/(MXMy)], ...(5) and Mx, My and Mxy refer to second order moments defined as (Fig,4 , step 5101):
Mx = Ff(x.Y)(x-xo)2.
My = Ff(x,y)(Y-Yo)2, and Mxy = Ff(x.Y)(x-xo)(Y-Yo) ....(6) The summation spans a circular area centered at (xo,yo) with a diameter of 8 pixels. ~ is defined as (Fig. 4 step 5102):
a = (Mxsin26+ Mycos29 + 2Mxysin6cos6)/(Mx+ My)...(7) where a ranges from 0 to 1.
Once the orientation is determined, comparison between the unknown document portion and the templates can be performed (Fig. 3, step S7). The templates are first rotated before matching in accordance with the determined orientation so that the axis of the anchor point aligns with the major direction of the block. The rotation is performed in a conventional manner by:
x' = xcos~+ y sink and y' _ -xsin~ + y cosh, ...(8) where (x, y) and (x', y') are the coordinates of a pixel before and after rotation, respectively and ~ is the determined angle.
Known image comparison means can be employed to compare the block with the anchor points on the templates. By way of example, Figure lA illustrates a template 2 containing a pre-chosen anchor point 4 with a known orientation 6. Figure 1 B discloses a scanned document 8 contained within a platen image containing an active block 10 with determined orientation 12. The orientation of template 2 is adjusted to match the orientation of unknown document 10. This is illustrated in Figure 1C. Finally, template 2 is matched with document 8, as is illustrated in Figure 1D.
The following equation may be used in matching:
r= Fvx tx ~v x,y Ft2(x,y) ...(9) where v(x,y) and t(x,y) are intensity values at (x,y) in the smoothed platen image and template image respectively, and the summation is over the size of the template. The matching is performed on all three channels.The value of r ranges from -1 to 1, wherein 1 indicates a perfect match. A
positive match is declared if r is greater than a preset threshold.
Should currency be discovered from a positive match between the template and the unknown document (Fig 3,step S8), the photocopier or printer 28 may be deactivated or the portion of the platen image containing the unknown document may be deleted from the final printed image (Fig. 3,step S9), and the operation is terminated (Fig.3, step S11). If no currency is discovered and there are more blocks to be examined, the operation returns to step S4 for the next block (Fig.3, step S10); otherwise, the operation is terminated (Fig. 3, step S11).
215'~7~1 As counterfeiting is generally an irregular occurrence, the probability of a negative matching result is far greater than that of a positive one. To save computation, the matching is performed hierarchically, from several points to the entire template and from a low resolution to a high resolution. (The "high resolution" is a relative term. It is typically about 16 pixels per inch.) This enables an abort of a test if a mismatch is perceived at an early stage. In most cases, fairly reliable results can be obtained at low resolutions. High resolution is merely used for final verification.
The edge orientation estimation also provides an estimated translation along the normal direction. Furthermore, template matching at low resolution is insensitive to small shifts. As a result, no translation compensation is necessary at low resolution. Should a high resolution matching be necessary, which is likely seldom, either a translation shift is estimated or the matching is performed for several different translations.
Applying more anchor points can increase the detection reliability. Roughly speaking, the miss detection rate is p" for n anchor points, where p is the miss detection rate for a single anchor point.
While this invention has been described in conjunction with specific embodiments thereof, it is evident that many alternatives, modifications and variations will be apparent to those skilled in the art.
Accordingly, the preferred embodiments of the invention as set forth herein are intended to be illustrative, not limiting. Various changes may be made without departing from the spirit and scope of the invention as defined in the following claims.
For example, if the currency to be detected are limited to a relatively small set, for example, U.S. dollars only, more sophisticated anchor points such as corner points can be used to increase the efficiency of the process. Other variations include different template matching techniques, different edge orientation estimation methods, and different schemes for combination of anchor points. Such variations may effect reliability, complexity, speed, and the constraints on the patterns to be detected.
BACKGROUND OF THE INVENTION
The present invention relates to an anti-counterfeit detector and method for identifying whether an image to be photocopied contains one or several pre-selected monetary note patterns.
Preventing colour copiers from being misused for counterfeiting has currently drawn more and more attention. In determining whether a colour copier is being used for counterfeiting, a detector compares a known currency image with an image begin copied. A problem arises in that it is difficult to detect the patterns in a rotation and shift invariant manner.
Specifically, the pattern could be of any orientation and at any location on the image. The orientation and the location of the note can be relatively simple to obtain in the case of a single note with a plain background. However, it is difficult to obtain orientation and location if multiple notes are involved and/or the notes are embedded in some complicated image background.
SUMMARY OF THE INVENTION
It is therefore an object of an aspect of the present invention to provide an anti-counterfeit pattern detector that detects patterns in a rotation and shift invariant manner.
It is a further object of an aspect of the present invention to provide greater accuracy and lower analysis time for the currency detection and orientation process.
Further aspects of the present invention are as follows:
An anti-counterfeit detector for detecting whether an image to be photocopies is a currency note, the detector comprising:
storage means for storing a plurality of templates, each of said plurality of templates comprising at least one predetermined anchor point;
r - la -examining means for examining a portion of said image to be photocopied and for determining whether said portion contains a predetermined pixel intensity arrangement;
orientation determining means for determining an orientation of said predetermined pixel intensity arrangement;
positioning means for positioning said template in alignment with said predetermined pixel intensity arrangement in accordance with said orientation determined by said orientation determining means; and matching means for comparing said plurality of templates to said predetermined pixel intensity arrangement;
wherein said predetermined pixel intensity arrangement is an edge and said examining means comprises means for calculating edge strength within said portion in accordance with the following equations:
f (x. Y~ - ~ d~ ~ 3 (x, Y~ + ~ dH ~ 3 (x, Y~ , and w = Ef (x, y) where d~ and dH are differences in intensity of neighbouring pixels in vertical and horizontal directions respectively, W is edge strength and summation is over said portion.
An anti-counterfeit detector for detecting whether an image to be photocopied is a currency note, the detector comprising:
a memory for storing a plurality of templates, each of said plurality of templates comprising at least one predetermined anchor point;
a scanner controller that examines a portion of said image to be photocopied and that determines whether said portion contains a predetermined pixel intensity arrangement;
an orientation detector operatively coupled to said scanner controller that detects an orientation of said portion;
a - lb -a positioning controller for aligning orientation of said template to said orientation of said portion; and a matching controller operatively coupled to said memory, said scanner controller, positioning controller and said orientation detector, said matching controller comparing said plurality of templates to said portion in accordance with said orientation;
wherein said predetermined pixel intensity arrangement is an edge and said examining means comprises means for calculating edge strength within said portion in accordance with the following equations:
f (x, Y) - ~ d~ ~ 3 (x, y) + ~ dH ~ 3 (x, y) , and w = Ef (x, y) where d~ and dH are differences in intensity of neighbouring pixels in vertical and horizontal directions respectively, W is edge strength and summation is over said portion.
An anti-counterfeit method for detecting whether an image to be photocopied is a currency note, the method comprising the steps of:
storing a plurality of templates, each of said plurality of templates comprising at least one predetermined anchor point;
examining a portion of said image to be photocopied and determining whether said portion contains a predetermined pixel intensity arrangement;
determining an orientation of said predetermined pixel intensity arrangement;
positioning said templates in alignment with said predetermined pixel intensity arrangement in accordance with said orientation determined by said orientation determination means; and comparing said plurality of templates to said predetermined pixel intensity arrangement;
A
wherein said predetermined pixel intensity arrangement comprises an edge and the step of examining comprises determining edge strength within said portion in accordance with the following equations:
f (x, Y) - ~ dV ~ 3 (x, y) + ~ dH ~ 3 (x, y) , and W = Ef (x, y) where d~ and dH are differences in intensity of neighbouring pixels in vertical and horizontal directions respectively, W is edge strength and summation is over said portion.
BRIEF DESCRIPTION OF THE DRAWINGS
These and other aspects of the present invention will become apparent from the following detailed description of the preferred embodiments when taking in conjunction with the accompanying drawings, in which:
A
2157'11 _2_ Figures lA-D illustrate matching a template to a detected sample;
Figure 2 is a block diagram of the system of the invention;
Figure 3 is a flowchart illustrating the operation in the invention; and Figure 4 is a flowchart illustrating orientation estimation according to the invention.
DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS
The following detailed description of preferred embodiments is applicable to numerous detection and imaging systems as would be contemplated by those of ordinary skill.
The invention will be described with general reference to Fig. 3 .
In order to detect whether a color copier is being used for counterfeiting, the detector 1 is first trained off-line with example notes.
A template is created by an initial sampling of currencies (hereinafter template) suspected of being photocopied and counterfeited.
Conventional sampling is preferably performed in a low resolution.
The training includes sampling the templates and selecting one or several anchor points. Only those pixels on relatively straight edges, preferably long straight edges, are qualified to be anchor points. Anchor points are used to determine the edge orientation of the unknown document or documents and orient the template accordingly (discussed below). There is no set limit as to the number of possible anchor points in each template. The greater the number of preselected anchor points, the higher the detection accuracy, but the more time of analysis.
One-dollar bills may be sampled at 16 dpi and 32 dpi, respectively. That resolution is sufficient to make a positive determination of the presence of currency when the template is compared with the image being photocopied. The template is stored in a computer memory 26 (Fig. 2) for future image comparison with the unknown document.
The detecting of currency notes placed on a platen will now be described. A digital color copier contains functionally a scanning part 20 and a printing part 28 as indicated in Fig.. 2. The currency detector 1 is placed in parallel to the normal video pass 3o. A data processor (CPU) 22 performs the function of the detector 1.
21~~711 When an image of the platen (platen image) containing the unknown document is scanned by the scanning part of the copier, the signal is also sent to the detector 1. The image is sampled using conventional means and at the resolution of the template (Fig. 3, step 51), and stored into the image buffer 21.
The information of a scanned color image is typically organized into three or four channels. The most commonly used sets of channels, or color spaces, are RGB and CIELAB. In RGB, three channels carry red (R), green (G), and blue (B) signals, respectively, while in CIELAB, L* channel represents luminance information, and a* and b* channels represent chromanance information. If RGB space is used, all three channels will be sampled at the same resolution of 16 dpi or 32 dpi. If L*a*b* space is used, a* and b* channels are sampled at a resolution that is half of that of the L*
channel. Specifically, L* is sampled at l6dpi or 32 dpi, and a* and b* are sampled at 8dpi or l6dpi.
After sampling, the color image of the platen is smoothed (Fig.
3, step S2). The process of smoothing entails the averaging of the value of pixels within a given area and reassigning the averaged value to the center pixel. For example, a block area of 3 x 3 pixels may contain pixels of different value. An average value is obtained for the nine pixels, and is assigned to the center pixel. The reassigned value produces a new version of the platen image providing less prominent fine texture patterns that could otherwise confuse edge detection and orientation estimation (see below).
The smoothed platen image is examined block by block, with a typical block size of 8 x 8 pixels. The blocks can be overlapping. Each block is examined by the data processor 22 (Fig. 2 to see if it possibly contains a pixel intensity orientation that corresponds to a preselected anchor point on the template. The "quiet blocks" containing little pixel variation, can be initially discarded as an edge is not present within the block (Fig. 3 , step 53). For remaining blocks, the orientation of the edges contained in the block is estimated. The edge strength and orientation estimation is only performed on one of the color channels of the smoothed scanned image.
In RGB space it is on G channel, and in CIELAB space it is on L* channel.
The currency images often contain many fine textures that could be mistaken as edges. It is therefore very important to reduce the influence of texture during the edge strength and orientation estimation.
The edge strength at a pixel (x,y) is quantified by the following formula:
f(x,Y) = Idvl3(x,Y) + IdHl3(x,Y) ...(1) ._ wl~'~711 where d~ and dH are differences in intensity of neighboring pixels of the smoothed image in the vertical and horizontal directions, respectively. The formula emphasizes strong differences, which typically characterize edges, more than the weak differences, which usually represent the texture.
The edge strength of the block is determined by the following equation:
W = Ff(x,y), ...(2) where the summation is over the 8x8 pixel block.
The determined edge strength W is measured against a pre-chosen threshold. (Fig. 3 step 54). A block with weighted edge strength in excess of the threshold contains a major edge. If the edge strength does not exceed the pre-chosen threshold, the block is discarded, and a subsequent block is examined (Fig.3,step S5).
If it is determined that a block contains an edge, before it can be compared (matched) with the stored templates, the orientation of the edge must be determined (Fig.3 , step S6).
Estimation of edge orientation (Fig.3, step S6) will now be described with reference to Fig. 4 , For the detector to properly determine counterfeiting, orientation estimation must be reliable and relatively accurate. In the present invention, a method based on second order moments is utilized, which is accurate and robust to noise.
First, a candidate for the anchor point is found by calculating xo = [Ef(x,y)x]/W, and Yo = IEf(x.Y)Y]~W, ...(3) where the summations are over the 8x8 block (Fig. 4 , step 5100). The orientation of the edge is then evaluated using the following equation (Fig. 4, step 5103):
~=8+0.5n, ifs>0.5;
8, otherwise, ...(4) where A = 0.5tan-~[2Mxy/(MXMy)], ...(5) and Mx, My and Mxy refer to second order moments defined as (Fig,4 , step 5101):
Mx = Ff(x.Y)(x-xo)2.
My = Ff(x,y)(Y-Yo)2, and Mxy = Ff(x.Y)(x-xo)(Y-Yo) ....(6) The summation spans a circular area centered at (xo,yo) with a diameter of 8 pixels. ~ is defined as (Fig. 4 step 5102):
a = (Mxsin26+ Mycos29 + 2Mxysin6cos6)/(Mx+ My)...(7) where a ranges from 0 to 1.
Once the orientation is determined, comparison between the unknown document portion and the templates can be performed (Fig. 3, step S7). The templates are first rotated before matching in accordance with the determined orientation so that the axis of the anchor point aligns with the major direction of the block. The rotation is performed in a conventional manner by:
x' = xcos~+ y sink and y' _ -xsin~ + y cosh, ...(8) where (x, y) and (x', y') are the coordinates of a pixel before and after rotation, respectively and ~ is the determined angle.
Known image comparison means can be employed to compare the block with the anchor points on the templates. By way of example, Figure lA illustrates a template 2 containing a pre-chosen anchor point 4 with a known orientation 6. Figure 1 B discloses a scanned document 8 contained within a platen image containing an active block 10 with determined orientation 12. The orientation of template 2 is adjusted to match the orientation of unknown document 10. This is illustrated in Figure 1C. Finally, template 2 is matched with document 8, as is illustrated in Figure 1D.
The following equation may be used in matching:
r= Fvx tx ~v x,y Ft2(x,y) ...(9) where v(x,y) and t(x,y) are intensity values at (x,y) in the smoothed platen image and template image respectively, and the summation is over the size of the template. The matching is performed on all three channels.The value of r ranges from -1 to 1, wherein 1 indicates a perfect match. A
positive match is declared if r is greater than a preset threshold.
Should currency be discovered from a positive match between the template and the unknown document (Fig 3,step S8), the photocopier or printer 28 may be deactivated or the portion of the platen image containing the unknown document may be deleted from the final printed image (Fig. 3,step S9), and the operation is terminated (Fig.3, step S11). If no currency is discovered and there are more blocks to be examined, the operation returns to step S4 for the next block (Fig.3, step S10); otherwise, the operation is terminated (Fig. 3, step S11).
215'~7~1 As counterfeiting is generally an irregular occurrence, the probability of a negative matching result is far greater than that of a positive one. To save computation, the matching is performed hierarchically, from several points to the entire template and from a low resolution to a high resolution. (The "high resolution" is a relative term. It is typically about 16 pixels per inch.) This enables an abort of a test if a mismatch is perceived at an early stage. In most cases, fairly reliable results can be obtained at low resolutions. High resolution is merely used for final verification.
The edge orientation estimation also provides an estimated translation along the normal direction. Furthermore, template matching at low resolution is insensitive to small shifts. As a result, no translation compensation is necessary at low resolution. Should a high resolution matching be necessary, which is likely seldom, either a translation shift is estimated or the matching is performed for several different translations.
Applying more anchor points can increase the detection reliability. Roughly speaking, the miss detection rate is p" for n anchor points, where p is the miss detection rate for a single anchor point.
While this invention has been described in conjunction with specific embodiments thereof, it is evident that many alternatives, modifications and variations will be apparent to those skilled in the art.
Accordingly, the preferred embodiments of the invention as set forth herein are intended to be illustrative, not limiting. Various changes may be made without departing from the spirit and scope of the invention as defined in the following claims.
For example, if the currency to be detected are limited to a relatively small set, for example, U.S. dollars only, more sophisticated anchor points such as corner points can be used to increase the efficiency of the process. Other variations include different template matching techniques, different edge orientation estimation methods, and different schemes for combination of anchor points. Such variations may effect reliability, complexity, speed, and the constraints on the patterns to be detected.
Claims (15)
1. An anti-counterfeit detector for detecting whether an image to be photocopies is a currency note, the detector comprising:
storage means for storing a plurality of templates, each of said plurality of templates comprising at least one predetermined anchor point;
examining means for examining a portion of said image to be photocopied and for determining whether said portion contains a predetermined pixel intensity arrangement;
orientation determining means for determining an orientation of said predetermined pixel intensity arrangement;
positioning means for positioning said template in alignment with said predetermined pixel intensity arrangement in accordance with said orientation determined by said orientation determining means; and matching means for comparing said plurality of templates to said predetermined pixel intensity arrangement;
wherein said predetermined pixel intensity arrangement is an edge and said examining means comprises means for calculating edge strength within said portion in accordance with the following equations:
f(x,y) = ~ (x,y) + ~ (x,y) , and W = .SIGMA.f(x,y) where d v and d H are differences in intensity of neighbouring pixels in vertical and horizontal directions respectively, W is edge strength and summation is over said portion.
storage means for storing a plurality of templates, each of said plurality of templates comprising at least one predetermined anchor point;
examining means for examining a portion of said image to be photocopied and for determining whether said portion contains a predetermined pixel intensity arrangement;
orientation determining means for determining an orientation of said predetermined pixel intensity arrangement;
positioning means for positioning said template in alignment with said predetermined pixel intensity arrangement in accordance with said orientation determined by said orientation determining means; and matching means for comparing said plurality of templates to said predetermined pixel intensity arrangement;
wherein said predetermined pixel intensity arrangement is an edge and said examining means comprises means for calculating edge strength within said portion in accordance with the following equations:
f(x,y) = ~ (x,y) + ~ (x,y) , and W = .SIGMA.f(x,y) where d v and d H are differences in intensity of neighbouring pixels in vertical and horizontal directions respectively, W is edge strength and summation is over said portion.
2. An anti-counterfeit detector according to claim 1 wherein if said edge strength does not exceed a predetermined threshold, said portion is discarded.
3. An anti-counterfeit detector according to claim 1, wherein said orientation determining means comprises means for determining edge orientation .PHI. in accordance with the following equations:
x0 = [.SIGMA.f(x,y)x]/W, and y0 = [.SIGMA.f(x,y)y]/W, wherein summation is over said portion .PHI. = .theta. + 0.5.pi., if .epsilon. ~ 0.5; and .PHI. = .theta. if .epsilon. < 0.5 wherein .PHI. = 0.5tan-1[2M xy/(M xM y)]
.epsilon. = (M xsin2.theta. + M ycos2.theta. + 2M xysin.theta.cos.theta.) / (M
x + M y) wherein M x, M y, and M xy refer to second order moments defined as M x = .SIGMA.f(x,y) (x - x0) 2 M y = .SIGMA.f(x,y) (y - Y0) 2 M zy = .SIGMA.f(x,y) (x - x0) (y - y0) , and summation is over a circular area with a preselected diameter and center at (x0, y0) .
x0 = [.SIGMA.f(x,y)x]/W, and y0 = [.SIGMA.f(x,y)y]/W, wherein summation is over said portion .PHI. = .theta. + 0.5.pi., if .epsilon. ~ 0.5; and .PHI. = .theta. if .epsilon. < 0.5 wherein .PHI. = 0.5tan-1[2M xy/(M xM y)]
.epsilon. = (M xsin2.theta. + M ycos2.theta. + 2M xysin.theta.cos.theta.) / (M
x + M y) wherein M x, M y, and M xy refer to second order moments defined as M x = .SIGMA.f(x,y) (x - x0) 2 M y = .SIGMA.f(x,y) (y - Y0) 2 M zy = .SIGMA.f(x,y) (x - x0) (y - y0) , and summation is over a circular area with a preselected diameter and center at (x0, y0) .
4. An anti-counterfeit detector according to claim 3 wherein said preselected diameter is 8 pixels.
5. An anti-counterfeit detector according to claim 4, wherein said positioning means comprises means for rotating said template according to the following equation:
x' = xcos.PHI. + ysin.PHI., and y' = -xsin.PHI. + ycos.PHI.
where (x,y) and (x',y') are coordinates of a pixel before and after rotation respectively.
x' = xcos.PHI. + ysin.PHI., and y' = -xsin.PHI. + ycos.PHI.
where (x,y) and (x',y') are coordinates of a pixel before and after rotation respectively.
6. An anti-counterfeit detector according to claim 1 wherein said matching means further comprises means for determining matching strength r in accordance with the following equation:
where v(x,y) and t(x,y) are intensity values at (x,y) in the image and each of the plurality of templates, respectively, and summation is over a template size.
where v(x,y) and t(x,y) are intensity values at (x,y) in the image and each of the plurality of templates, respectively, and summation is over a template size.
7. An anti-counterfeit detector according to claim 6 wherein a match is indicated by r being greater than a preset threshold.
8. An anti-counterfeit detector for detecting whether an image to be photocopied is a currency note, the detector comprising:
a memory for storing a plurality of templates, each of said plurality of templates comprising at least one predetermined anchor point;
a scanner controller that examines a portion of said image to be photocopied and that determines whether said portion contains a predetermined pixel intensity arrangement;
an orientation detector operatively coupled to said scanner controller that detects an orientation of said portion;
a positioning controller for aligning orientation of said template to said orientation of said portion; and a matching controller operatively coupled to said memory, said scanner controller, positioning controller and said orientation detector, said matching controller comparing said plurality of templates to said portion in accordance with said orientation;
wherein said predetermined pixel intensity arrangement is an edge and said examining means comprises means for calculating edge strength within said portion in accordance with the following equations:
and where d v and d H are differences in intensity of neighbouring pixels in vertical and horizontal directions respectively, W is edge strength and summation is over said portion.
a memory for storing a plurality of templates, each of said plurality of templates comprising at least one predetermined anchor point;
a scanner controller that examines a portion of said image to be photocopied and that determines whether said portion contains a predetermined pixel intensity arrangement;
an orientation detector operatively coupled to said scanner controller that detects an orientation of said portion;
a positioning controller for aligning orientation of said template to said orientation of said portion; and a matching controller operatively coupled to said memory, said scanner controller, positioning controller and said orientation detector, said matching controller comparing said plurality of templates to said portion in accordance with said orientation;
wherein said predetermined pixel intensity arrangement is an edge and said examining means comprises means for calculating edge strength within said portion in accordance with the following equations:
and where d v and d H are differences in intensity of neighbouring pixels in vertical and horizontal directions respectively, W is edge strength and summation is over said portion.
9. An anti-counterfeit method for detecting whether an image to be photocopied is a currency note, the method comprising the steps of:
storing a plurality of templates, each of said plurality of templates comprising at least one predetermined anchor point;
examining a portion of said image to be photocopied and determining whether said portion contains a predetermined pixel intensity arrangement;
determining an orientation of said predetermined pixel intensity arrangement;
positioning said templates in alignment with said predetermined pixel intensity arrangement in accordance with said orientation determined by said orientation determination means; and comparing said plurality of templates to said predetermined pixel intensity arrangement;
wherein said predetermined pixel intensity arrangement comprises an edge and the step of examining comprises determining edge strength within said portion in accordance with the following equations:
and where d v and d H are differences in intensity of neighbouring pixels in vertical and horizontal directions respectively, W is edge strength and summation is over said portion.
storing a plurality of templates, each of said plurality of templates comprising at least one predetermined anchor point;
examining a portion of said image to be photocopied and determining whether said portion contains a predetermined pixel intensity arrangement;
determining an orientation of said predetermined pixel intensity arrangement;
positioning said templates in alignment with said predetermined pixel intensity arrangement in accordance with said orientation determined by said orientation determination means; and comparing said plurality of templates to said predetermined pixel intensity arrangement;
wherein said predetermined pixel intensity arrangement comprises an edge and the step of examining comprises determining edge strength within said portion in accordance with the following equations:
and where d v and d H are differences in intensity of neighbouring pixels in vertical and horizontal directions respectively, W is edge strength and summation is over said portion.
10. The method according to claim 9, wherein an examined portion is discarded if its edge strength does not exceed a preselected threshold.
11. The method according to claim 9, wherein the step of determining comprises determining edge orientation .PHI. defined in accordance with the following equations:
x o = [.SIGMA.f(x,y)x]/W, and y o = [.SIGMA.f (x,y)y]/W, wherein summation is over said portion .PHI. = .theta. + 0.5II, if .epsilon. ~ 0.5; and .PHI. = .theta. if .epsilon. < 0.5 wherein .theta. = 0.5tan-1 [2M xy / (M x M y)]
.epsilon. = (M x sin2.theta. + M y cos 2 .theta. + 2M xy sin .theta.cos.theta.) / (M x + M y) wherein M x, M y, and M xy refer to second order moments defined as M x = .SIGMA.f (x, y) (x - x o)2 M y = .SIGMA.f (x, y) (y - y o)2 M zy = .SIGMA.f (x, y) (x - x o) (y - y o), and summation is over a circular area of preselected diameter centered at (x o, y o).
x o = [.SIGMA.f(x,y)x]/W, and y o = [.SIGMA.f (x,y)y]/W, wherein summation is over said portion .PHI. = .theta. + 0.5II, if .epsilon. ~ 0.5; and .PHI. = .theta. if .epsilon. < 0.5 wherein .theta. = 0.5tan-1 [2M xy / (M x M y)]
.epsilon. = (M x sin2.theta. + M y cos 2 .theta. + 2M xy sin .theta.cos.theta.) / (M x + M y) wherein M x, M y, and M xy refer to second order moments defined as M x = .SIGMA.f (x, y) (x - x o)2 M y = .SIGMA.f (x, y) (y - y o)2 M zy = .SIGMA.f (x, y) (x - x o) (y - y o), and summation is over a circular area of preselected diameter centered at (x o, y o).
12. The method according to claim 11, wherein said preselected diameter is 8 pixels.
13. The method according to claim 11, wherein the step of aligning comprises rotation of said templates according to the following equations:
x' = xcos.PHI. + ysin.PHI., and y' = -xsin.PHI. + ycos.PHI.
where (x, y) and (x',y') are coordinates of a pixel before and after rotation respectively.
x' = xcos.PHI. + ysin.PHI., and y' = -xsin.PHI. + ycos.PHI.
where (x, y) and (x',y') are coordinates of a pixel before and after rotation respectively.
14. The method according to claim 9, wherein the step of comparing comprises determining matching strength r in accordance with the following equation:
where v(x,y) and t(x,y) are intensity values at (x, y) in a smoothed plate image and template image respectively, and summation is over template size.
where v(x,y) and t(x,y) are intensity values at (x, y) in a smoothed plate image and template image respectively, and summation is over template size.
15. The method according to claim 14 wherein a match is indicated by r being greater than a preset threshold.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US08/323,652 US5533144A (en) | 1994-10-17 | 1994-10-17 | Anti-counterfeit pattern detector and method |
| US08/323,652 | 1994-10-17 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CA2157711A1 CA2157711A1 (en) | 1996-04-18 |
| CA2157711C true CA2157711C (en) | 1999-12-21 |
Family
ID=23260129
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA002157711A Expired - Fee Related CA2157711C (en) | 1994-10-17 | 1995-09-07 | Anti-counterfeit pattern detector and method |
Country Status (6)
| Country | Link |
|---|---|
| US (1) | US5533144A (en) |
| EP (1) | EP0708411B1 (en) |
| JP (1) | JP3825070B2 (en) |
| BR (1) | BR9504401B1 (en) |
| CA (1) | CA2157711C (en) |
| DE (1) | DE69529324T2 (en) |
Families Citing this family (45)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6345104B1 (en) | 1994-03-17 | 2002-02-05 | Digimarc Corporation | Digital watermarks and methods for security documents |
| US6549638B2 (en) | 1998-11-03 | 2003-04-15 | Digimarc Corporation | Methods for evidencing illicit use of a computer system or device |
| US6449377B1 (en) | 1995-05-08 | 2002-09-10 | Digimarc Corporation | Methods and systems for watermark processing of line art images |
| US7113615B2 (en) | 1993-11-18 | 2006-09-26 | Digimarc Corporation | Watermark embedder and reader |
| US7286684B2 (en) * | 1994-03-17 | 2007-10-23 | Digimarc Corporation | Secure document design carrying auxiliary machine readable information |
| EP0765748A3 (en) * | 1995-09-29 | 1997-08-13 | Goss Graphics Systems Inc | Device for alignment of images in a control system for a printing press |
| US6104834A (en) | 1996-08-01 | 2000-08-15 | Ricoh Company Limited | Matching CCITT compressed document images |
| US5892854A (en) * | 1997-01-21 | 1999-04-06 | Xerox Corporation | Automatic image registration using binary moments |
| GB9703191D0 (en) * | 1997-02-15 | 1997-04-02 | Ncr Int Inc | Method and apparatus for screening documents |
| US6181813B1 (en) * | 1997-09-29 | 2001-01-30 | Xerox Corporation | Method for counterfeit currency detection using orthogonal line comparison |
| US6067374A (en) * | 1997-11-13 | 2000-05-23 | Xerox Corporation | Seal detection system and method |
| US6026186A (en) * | 1997-11-17 | 2000-02-15 | Xerox Corporation | Line and curve detection using local information |
| US6272245B1 (en) | 1998-01-23 | 2001-08-07 | Seiko Epson Corporation | Apparatus and method for pattern recognition |
| US6400834B1 (en) * | 1998-06-10 | 2002-06-04 | Micron Electonics, Inc. | Method for detecting photocopied or laser-printed documents |
| US6571001B2 (en) | 1998-06-10 | 2003-05-27 | Micron Technology, Inc. | System for detecting photocopied or laser-printed documents |
| US8290202B2 (en) | 1998-11-03 | 2012-10-16 | Digimarc Corporation | Methods utilizing steganography |
| JP3695217B2 (en) * | 1999-04-30 | 2005-09-14 | オムロン株式会社 | Image processing apparatus and image input apparatus |
| US6370271B2 (en) | 1999-04-30 | 2002-04-09 | Seiko Epson Corporation | Image processing apparatus and methods for pattern recognition |
| US6542629B1 (en) * | 1999-07-22 | 2003-04-01 | Xerox Corporation | Digital imaging method and apparatus for detection of document security marks |
| US7817844B2 (en) * | 1999-08-26 | 2010-10-19 | Nanogeometry Research Inc. | Pattern inspection apparatus and method |
| US7796801B2 (en) * | 1999-08-26 | 2010-09-14 | Nanogeometry Research Inc. | Pattern inspection apparatus and method |
| US6553136B1 (en) * | 1999-10-28 | 2003-04-22 | Hewlett-Packard Company | System and method for counterfeit protection |
| US7193744B2 (en) * | 2000-01-31 | 2007-03-20 | Canon Kabushiki Kaisha | Image forming system and its control method |
| US7002710B1 (en) | 2000-04-10 | 2006-02-21 | Hewlett-Packard Development Company, L.P. | High reliability forensic color marking system |
| US7002704B1 (en) | 2000-11-06 | 2006-02-21 | Xerox Corporation | Method and apparatus for implementing anti-counterfeiting measures in personal computer-based digital color printers |
| US6868174B2 (en) * | 2000-11-29 | 2005-03-15 | Xerox Corporation | Anti-counterfeit detection for low end products |
| US6470168B1 (en) | 2001-02-22 | 2002-10-22 | Xerox Corporation | ACD using an added low resolution CCD to prevent even partial printout |
| JP2003099788A (en) * | 2001-09-21 | 2003-04-04 | Sharp Corp | Image processor |
| US7334225B2 (en) * | 2003-04-28 | 2008-02-19 | International Business Machines Corporation | Method, system, and computer program product for on demand enablement of dormant computing resources |
| US20040260775A1 (en) * | 2003-06-20 | 2004-12-23 | Xerox Corporation | System and method for sending messages |
| US7302316B2 (en) * | 2004-09-14 | 2007-11-27 | Brigham Young University | Programmable autopilot system for autonomous flight of unmanned aerial vehicles |
| EP1969525A1 (en) * | 2005-12-23 | 2008-09-17 | Ingenia Holdings (UK)Limited | Optical authentication |
| US8527285B2 (en) * | 2006-06-28 | 2013-09-03 | Pitney Bowes Inc. | Postage printing system for printing both postal and non-postal documents |
| US7916924B2 (en) * | 2006-09-19 | 2011-03-29 | Primax Electronics Ltd. | Color processing method for identification of areas within an image corresponding to monetary banknotes |
| US7885450B2 (en) * | 2006-09-20 | 2011-02-08 | Primax Electronics Ltd. | Method for characterizing texture of areas within an image corresponding to monetary banknotes |
| US7706592B2 (en) * | 2006-09-20 | 2010-04-27 | Primax Electronics Ltd. | Method for detecting a boundary of a monetary banknote within an image |
| US7738690B2 (en) * | 2006-09-20 | 2010-06-15 | Primax Electronics Ltd. | Verification method for determining areas within an image corresponding to monetary banknotes |
| US7706593B2 (en) * | 2006-09-20 | 2010-04-27 | Primax Electronics Ltd. | Verification method for determining areas within an image corresponding to monetary banknotes |
| JP4743065B2 (en) * | 2006-09-27 | 2011-08-10 | ブラザー工業株式会社 | Image recognition apparatus, copying apparatus, and image recognition method |
| DE102007019107A1 (en) * | 2007-04-23 | 2008-10-30 | Giesecke & Devrient Gmbh | Method and device for checking value documents |
| US8150140B2 (en) * | 2008-12-22 | 2012-04-03 | Ngr Inc. | System and method for a semiconductor lithographic process control using statistical information in defect identification |
| JP5429869B2 (en) * | 2008-12-22 | 2014-02-26 | 株式会社 Ｎｇｒ | Pattern inspection apparatus and method |
| US9076195B2 (en) * | 2013-08-29 | 2015-07-07 | The Boeing Company | Methods and apparatus to identify components from images of the components |
| EP3365859A1 (en) * | 2015-10-20 | 2018-08-29 | Innogy Innovation Gmbh | Method and apparatus for creating a supplier data set containing information about a service supplier |
| WO2018190778A1 (en) | 2017-04-12 | 2018-10-18 | Thammasat University | Inspection procedure of the authenticity of brand-named leather bag |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US3885229A (en) * | 1972-10-28 | 1975-05-20 | Nippon Electric Co | Document scanning apparatus |
| JPS5313840A (en) * | 1976-07-23 | 1978-02-07 | Hitachi Ltd | Analogy calculator |
| SG102585A1 (en) * | 1989-02-10 | 2004-03-26 | Canon Kk | Apparatus for image reading or processing |
| EP0569982B1 (en) * | 1992-05-14 | 1999-03-10 | United Parcel Service Of America, Inc. | Method and apparatus for processing images with symbols with dense edges |
-
1994
- 1994-10-17 US US08/323,652 patent/US5533144A/en not_active Expired - Lifetime
-
1995
- 1995-09-07 CA CA002157711A patent/CA2157711C/en not_active Expired - Fee Related
- 1995-10-09 JP JP26123095A patent/JP3825070B2/en not_active Expired - Fee Related
- 1995-10-13 BR BRPI9504401-9A patent/BR9504401B1/en not_active IP Right Cessation
- 1995-10-17 EP EP95307359A patent/EP0708411B1/en not_active Expired - Lifetime
- 1995-10-17 DE DE69529324T patent/DE69529324T2/en not_active Expired - Lifetime
Also Published As
| Publication number | Publication date |
|---|---|
| BR9504401A (en) | 1997-05-27 |
| JP3825070B2 (en) | 2006-09-20 |
| US5533144A (en) | 1996-07-02 |
| CA2157711A1 (en) | 1996-04-18 |
| BR9504401B1 (en) | 2010-10-05 |
| DE69529324D1 (en) | 2003-02-13 |
| EP0708411B1 (en) | 2003-01-08 |
| JPH08185095A (en) | 1996-07-16 |
| DE69529324T2 (en) | 2003-08-28 |
| EP0708411A3 (en) | 1997-05-28 |
| EP0708411A2 (en) | 1996-04-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CA2157711C (en) | Anti-counterfeit pattern detector and method | |
| US6574366B1 (en) | Line and curve detection using local information | |
| US6272245B1 (en) | Apparatus and method for pattern recognition | |
| US6181813B1 (en) | Method for counterfeit currency detection using orthogonal line comparison | |
| US6343204B1 (en) | Detection and deterrence of counterfeiting of documents with tokens characteristic color and spacing | |
| US6067374A (en) | Seal detection system and method | |
| US5530772A (en) | Apparatus and method for testing bank notes for genuineness using Fourier transform analysis | |
| US6370271B2 (en) | Image processing apparatus and methods for pattern recognition | |
| EP0940780B1 (en) | Image processing device and method for detecting a reference pattern | |
| JP3689436B2 (en) | Anti-duplication method and anti-duplication device | |
| EP1953710B1 (en) | Counterfeit Deterrence Using Dispersed Miniature Security Marks | |
| KR19990006591A (en) | Image detection method, image detection device, image processing method, image processing device and medium | |
| US6731784B2 (en) | Detection and deterrence of counterfeiting of documents with a seal having characteristic color, size, shape and radial density profile | |
| JPS60229572A (en) | Picture processor | |
| US20080069424A1 (en) | Method for characterizing texture of areas within an image corresponding to monetary banknotes | |
| US7155051B2 (en) | Image recognition apparatus, image recognition method and image recognition program for specific pattern | |
| JP4014070B2 (en) | Pattern detection method and image processing control method | |
| JPH08115021A (en) | Color image processor | |
| JPH06251128A (en) | Image processor and copying machine using its processor | |
| JPH11110545A (en) | Image recognition device | |
| JP2000295470A (en) | Image output unit | |
| JPH0662235A (en) | Special original discriminating device | |
| JPH0774943A (en) | Image forming device | |
| JPH06217131A (en) | Special original discriminating method and copying machine with special original discriminating function | |
| JPH0662236A (en) | Special original discriminating device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| EEER | Examination request | ||
| MKLA | Lapsed |
Effective date: 20150908 |