JP6862683B2 - Judgment device and program - Google Patents

Description

The present invention relates to a technique for determining a printing method of an object.

In recent years, identification cards such as driver's licenses, basic resident register cards, residence cards, special permanent resident certificates, and passports that require identity verification may be in the form of cards. Since these ID cards are used for identity verification, it is necessary to strictly determine whether or not the ID card itself is genuine. Then, in determining the authenticity of such an identification card, as disclosed in Patent Documents 1 and 2, a method of determining the authenticity using an image obtained by imaging the surface of the identification card is known. Has been done.

The method described in Patent Document 1 determines the authenticity by utilizing the fact that the fake identification card is often manufactured by so-called area gradation printing by, for example, a laser printer or an inkjet printer. .. More specifically, in Patent Document 1, first, a specific area is extracted from the image of the entire ID card based on a preset format peculiar to the ID card. After that, when moire fringes (that is, an image pattern having a predetermined pattern arrangement) generated in the fake product manufactured by printing in the area gradation method is detected from the extracted area, the identification card is determined to be the fake product. judge.

Japanese Unexamined Patent Publication No. 2011-34535 Japanese Unexamined Patent Publication No. 2016-76079

However, in the technique described in Patent Document 1, moire fringes do not occur unless the image acquisition resolution is about the same as the minimum print resolution. Further, if the print resolution of the target printing device does not match the image acquisition device, the determination cannot be made.

An object of the present invention is to provide a determination device capable of stably determining whether or not printing is performed by an area gradation method without being affected by a change in brightness of a printed surface.

From one aspect of the present invention, the determination device includes an acquisition means for acquiring captured image of a part of the captured image of the object, a color component generating means for generating a color component of the captured image from the captured data, and the like. An area gradation method is printed on a part of the object based on the color flatness calculation means for determining the color flatness of the captured image based on the color component and the determined color flatness. The mask processing means includes a print determination means for determining whether or not the image is present, and a mask processing means for performing mask processing on a character area included in the captured image, and the mask processing means includes a plurality of pixels included in the captured image. Among them, a pixel having a brightness component of the first predetermined value or more and a pixel having a brightness component of the second predetermined value or less are detected, the area of the pixel and the peripheral pixels of the pixel is masked as the character area, and the color component generating means. Generates the color component for an area other than the character area masked by the masking process.

The above-mentioned determination device acquires the captured data of a part of the captured image of the object such as an identification card, and generates the color component of the captured image from the captured data. Then, the color flatness of the captured image is determined based on the color component, and based on the determined color flatness, it is determined whether or not a part of the object is printed by the area gradation method. Further, the determination device includes a mask processing means for performing mask processing on the character area included in the captured image . The mask processing means detects a pixel having a brightness component of the first predetermined value or more and a pixel having a brightness component of the second predetermined value or less among a plurality of pixels included in the captured image, and of the pixel and peripheral pixels of the pixel. The area is masked as the character area. The color component generating means generates the color component for a region other than the character region masked by the masking process. As a result, even if the captured image includes a character area, the determination can be made correctly.

In one aspect of the determination device, the color flatness calculating means calculates the color flatness based on any of the mean deviation, dispersion, and standard deviation of the color components in the plurality of pixel regions constituting the captured image. To do.

In another aspect of the above-mentioned determination device, the color flatness calculating means performs frequency analysis of color components in a plurality of pixel regions constituting the captured image, and is based on the magnitude of frequency components other than the DC component. The color flatness is calculated.

In another aspect of the above-mentioned determination device, the color flatness calculating means calculates the color flatness for each of a plurality of pixel regions constituting the captured image, and is a representative value of the obtained plurality of color flatnesses. Is determined, and the print determination means determines whether or not the area gradation type printing is applied to a part of the object based on the representative value. In this aspect, even when a region other than a single color is used as a part of the object, or when the surface of the object has scratches, stains, etc., the determination can be made correctly. Preferably, the color flatness calculating means is a standard based on the average value of the plurality of color flatnesses, the peak value in the histogram of the plurality of color flatnesses, and the average value of the plurality of color flatnesses. Any one of the average values of color flatness within the deviation range is used as the representative value.

In another aspect of the above-mentioned determination device, the imaging data is RGB data, and the color component generating means generates a color difference component from the RGB data and uses it as the color component. In this aspect, the color component can be generated by the calculation of the color difference component.

Another aspect of the above-mentioned determination device includes an authenticity determination means for determining an object as a fake when a part of the object is printed in an area gradation system. In this aspect, it is possible to determine whether the object is genuine or fake based on whether or not the area gradation printing is applied.

In another aspect of the present invention, a program executed by a determination device including a computer generates a color component of the captured image from the captured data, an acquisition means for acquiring captured data of a part of the captured image of the object. Color component generation means to be used, color flatness calculation means for determining the color flatness of the captured image based on the color component, and an area gradation method for a part of the object based on the determined color flatness. The computer is made to function as a print determining means for determining whether or not printing is performed, and a mask processing means for performing mask processing on a character area included in the captured data, and the mask processing means is the captured image. Among the plurality of pixels included in the above, a pixel having a brightness component of the first predetermined value or more and a pixel having a brightness component of the second predetermined value or less are detected, and the area of the pixel and the peripheral pixels of the pixel is masked as the character area. Then, the color component generating means generates the color component in a region other than the character region masked by the masking process. By executing this program on a computer, the above-mentioned determination device can be realized.

It is a figure explaining the basic principle of the determination method which concerns on embodiment. It is a block diagram which shows the structure of the determination apparatus which concerns on this embodiment. An example of the calculation method of the flatness judgment value is shown. It is explanatory drawing of the determination method in the determination area which is not a single color. It is explanatory drawing of the determination method of the representative value in the determination area which is not a single color. It is explanatory drawing of the mask processing of a character area. It is a flowchart of a determination process.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
[Basic principle]
First, the basic principle of the determination method according to the present embodiment will be described. The determination device of the present embodiment determines whether or not the printed portion of the object to be determined (hereinafter referred to as "object") such as an identification card is based on the area gradation method, and certifies the identity. Judge the authenticity of books. Genuine identification is printed by the sublimation transfer method. On the other hand, a fake ID card is often manufactured by so-called area gradation printing using image data obtained by imaging a real ID card with a color scanner. Therefore, the determination device of the present embodiment determines whether or not the printed surface of the object such as the identification card is printed by the area gradation method, and when the printed surface is formed by the area gradation method. , Judge the object as a fake. The area gradation type printing includes, for example, dot matrix type printing, printing by a laser printer or an inkjet printer, and the like.

When the object is a fake produced by area gradation printing, color variation occurs in the area painted in a single color such as the background of a face photograph included in an identification card. FIG. 1A shows a method of color expression by an area gradation printer. In the area gradation method, color printing is performed by adjusting the area for printing cyan, magenta, and yellow in the target unit printing area 31. For example, by printing cyan, magenta, and yellow in an area of 1/3 each in the unit print area 31, gray is printed as a whole. Therefore, the color variation is not felt when viewed at a resolution lower than the print resolution, but the color variation is recognized when viewed at a resolution higher than the print resolution.

FIG. 1B schematically shows a state in which a printed surface printed with gray by an area gradation method is read by a scanner having a resolution higher than the print resolution. Since the resolution of the scanner is higher than the print resolution, the unit reading area 32 of the scanner is smaller than the unit print area 31 in the area gradation type printing. Therefore, in the unit reading area of the scanner, the area ratio of the cyan, magenta, and yellow areas is not reduced to 1/3 each, and color bias occurs. In this way, when the printed surface printed by the area gradation method is read by a high-resolution scanner, color variations can be observed. On the other hand, when the object is genuine, such color variation does not occur. Therefore, in the present embodiment, whether or not the object is printed by the area gradation method by scanning the printed surface of the object with a high-resolution scanner and analyzing the color variation, that is, whether it is a fake or not. Judge whether or not.

Here, the present embodiment is particularly characterized in that the imaging data obtained by scanning with a scanner is divided into a brightness component and a color component, and a determination is made based on the color component. As described above, it is possible to determine whether or not the printing is based on the area gradation method by analyzing the color variation in the image data of the printing surface of the object, but depending on the color of the printing surface, it may be affected by the change in brightness. Judgment accuracy may decrease. For example, in the blue area used for a driver's license, the change in brightness is small and its influence is difficult to appear in the judgment result, but in the gold area, the change in brightness is large and the change in brightness is judged. The impact on the results will be greater. Therefore, in the present embodiment, the imaging data of the object is divided into a brightness component and a color component, and the color variation is analyzed based on the color component to reduce the influence of the brightness change in the determination.

Specifically, in the present embodiment, the luminance component is used as the brightness component, and the color difference component is used as the color component. That is, the luminance component and the color difference component are extracted from the image data taken by the scanner, and the color variation is detected based on the color difference component. As an index indicating color variation, "color difference flatness", which is the flatness of the color difference component, is used. That is, in the present embodiment, it is determined whether or not the printed surface is in the area gradation method based on the color difference flatness.

[Configuration of judgment device]
FIG. 2 is a block diagram showing a configuration of a determination device according to the present embodiment. The determination device 100 includes a processing unit 1, a scanner 2, a storage unit 3, an operation unit 4, a display unit 5, and an interface (I / F) 6. The determination device 100 reads an image of the printed surface of the identification card C, which is an example of the object, determines whether or not printing is performed by the area gradation method, and based on the result, the identification card is genuine. Judgment as to whether it is present or fake (authenticity judgment) is performed.

The processing unit 1 determines whether or not the printed surface is printed by the area gradation method based on the imaged data of the identification card C, and determines the authenticity. Specifically, the processing unit 1 includes a conversion unit 11, a flatness determination unit 12, and an authenticity determination unit 13. The processing unit 1 is actually composed of a computer such as a CPU.

The scanner 2 reads the printed surface of the identification card C and generates the imaging data thereof. The scanner 2 includes a reading unit 21, a specific unit 22, and an imaging unit 23. The imaging data generated by the scanner 2 is sent to the processing unit 1 via the interface 6.

The storage unit 3 stores a processing program executed by the processing unit 1 and data necessary for the processing. In addition, the storage unit 3 temporarily stores the imaging data generated by the scanner 2. The operation unit 4 is an input device operated by the user. The display unit 5 is a display that displays the result of the authenticity determination executed by the processing unit 1.

Next, the operation of the determination device 100 will be described. In the identification card C, type data indicating the type of the identification card is stored in an internal memory or the like. Specifically, the type data indicates whether the identification card is a driver's license, a passport, or the like. When the identification card C is set in the scanner 2, the reading unit 21 of the scanner 2 acquires the type data from the identification card C by contact or non-contact communication and sends it to the specific unit 22.

The specific unit 22 stores a predetermined determination area for each type of certificate. The determination area is information indicating which part of each certificate should be read by the scanner 2, and is basically information indicating a color-printed part of each certificate. For example, in the case of a driver's license, the determination area is set in the owner's photo column and / or the expiration date column (green, blue or gold). When the identification unit 22 acquires the type data from the reading unit 21, it identifies a determination area determined in advance for each type of certificate and sends it to the imaging unit 23. The imaging unit 23 reads the corresponding range of the identification card C based on the determination area designated by the specific unit 22, and generates imaging data which is RGB data. Then, the imaging unit 23 sends the generated imaging data to the processing unit 1 via the interface 6.

In the processing unit 1, the imaging data is supplied to the conversion unit 11. The conversion unit 11 separates the imaging data, which is RGB data, into a luminance component and a color difference component. Specifically, the conversion unit 11 performs YUV conversion on the RGB data to generate the luminance component Y and the color difference components U and V. Then, the conversion unit 11 supplies the generated color difference components U and V to the flatness determination unit 12.

The flatness determination unit 12 obtains the color difference flatness based on the color difference components U and V, and supplies the obtained color difference flatness value to the authenticity determination unit 13 as the color difference flatness determination value. The authenticity determination unit 13 determines whether the identification card is genuine or fake based on the color difference flatness determination value. This determination result is sent to the display unit 5 and displayed.

In the above configuration, the scanner 2 is an example of the acquisition means of the present invention, the conversion unit 11 is an example of the color component generation means of the present invention, and the flatness determination unit 12 is the color flatness calculation means and printing of the present invention. The authenticity determination unit 13 is an example of the authenticity determination means of the present invention, and the processing unit 1 is an example of the mask means of the present invention.

[Color difference flatness judgment]
Next, the color difference flatness determination performed by the flatness determination unit 12 will be described in detail. The color difference flatness determination can be performed by any of the following methods.
(1) Color difference flatness determination based on average value difference In one method, the flatness determination unit 12 relates to a region of m × n pixels (hereinafter, also referred to as “determination pixel block”) in an image of color difference components U and V. The difference value between the pixel value of each pixel and the average value of the pixel values of all the pixels in the determination pixel block is used as the color difference flatness determination value. Specifically, as the color difference flatness determination value, mean deviation, variance, standard deviation, or the like can be used. In this case, the higher the color difference flatness, the smaller the color difference flatness determination value.

When the average deviation is used, the color difference flatness determination value is the absolute value of the difference between each pixel value in each determination pixel block of the color difference components U and V and the average value of all pixel values in the determination pixel block. Given as an average. Now, in the determination pixel block, each pixel value of the color difference component U is set to "U _mn ", the average value of all the pixel values of the color difference component U is set to "μ _u ", and each pixel value of the color difference component V is set to "V _mn ". _Assuming that the average value of all the pixel values of the color difference component V is "μ _v ", the flatness determination unit 12 sets the average deviation obtained by the following equation as the color difference flatness determination values "U Evval" and "V _Evval ". ..

When dispersion is used, the color difference flatness determination value is the squared average of the difference between each pixel value in each determination pixel block of the color difference components U and V and the average value of all pixel values in the determination pixel block. Given. Now, in the determination pixel block, each pixel value of the color difference component U is set to "U _mn ", the average value of all the pixel values of the color difference component U is set to "μ _u ", and each pixel value of the color difference component V is set to "V _mn ". _Assuming that the average value of all the pixel values of the color difference component V is "μ _v ", the flatness determination unit 12 sets the variance obtained by the following equation as the color difference flatness determination values "U Evval" and "V _Evval ".

When the standard deviation is used, the flatness determination unit 12 sets the square root of the variance obtained by the above equations (3) and (4) as the color difference flatness determination values "U _Evval " and "V _Evval ".

(2) Judgment of color difference flatness by frequency analysis Another method is to perform frequency decomposition such as DCT (discrete cosine transform) and Hadamard transform on the judgment pixel block of m × n pixels, and color difference from values other than the DC value. Obtain the flatness judgment value. For example, when the determination pixel block is an area of 4 × 4 pixels and the Hadamard transform is used, the conversion result shown in FIG. 3 (B) can be obtained by using the conversion formula shown in FIG. 3 (A). In the matrix shown in FIG. 3B, the DC component (a value four times the average value of 4 × 4 pixels) is shown at the position (1,1), and the coefficient of each frequency component is shown at the other positions. Is shown. The flatness determination unit 12 uses the squared average of coefficients other than the DC component as the color difference flatness determination value. When the color variation in the judgment pixel block of the identification card C is large, the value of each coefficient other than the DC component becomes large, so that the squared average value becomes large and the color difference flatness judgment value becomes large. .. On the other hand, when the color variation in the judgment pixel block of the identification card C is small, the values of the respective coefficients other than the DC component are small, so that the squared average value is small, and the color difference flatness judgment value is It becomes smaller. Therefore, the higher the color difference flatness, the smaller the color difference flatness determination value.

FIG. 3C shows the results of determining the color difference flatness using the Hadamard transform for a plurality of samples including genuine and fake samples. The horizontal axis U indicates the color difference flatness determination value of the color difference component U, and the vertical axis V indicates the color difference flatness determination value of the color difference component V. As shown in the figure, the real color difference flatness determination values are all small and gathered at about "2" or less, whereas the fake color difference flatness determination values are large and the values vary widely. Therefore, by setting a predetermined threshold value for the color difference flatness determination value, it is possible to determine whether the product is genuine or fake based on the color difference flatness determination value.

[Judgment area of object]
Next, the determination region, which is the range in which the imaging data is generated in the object, will be described. In the above description, it is assumed that the determination area is basically a single color, but an area other than a single color can be used as the determination area. For example, as shown in FIG. 4A, in the driver's license which is an example of the identification card C, the area AR shown by the broken line is set as the determination area, and the scanner 2 generates the imaging data of this determination area. The processing unit 1 obtains a color difference flatness determination value for each determination pixel block or a predetermined pixel block larger than the determination pixel block by using the above-mentioned average value difference or frequency analysis, and represents from the obtained plurality of color difference flatness determination values. A value is determined, and authenticity is determined using this representative value.

Here, a method of determining the representative value will be described. FIG. 5 shows each of a plurality of determination pixel blocks (areas of m × m pixels) belonging to the determination area based on the imaging data generated using an area other than a single color as the determination area, such as the area AR of FIG. 4 (A). It is a histogram of the color difference flatness judgment value calculated in. In this case, since the color difference flatness obtained in a single color region such as the background region in the region AR is high, many small color difference flatness determination values can be obtained. On the other hand, since the color difference flatness is low in the boundary portion between the head and the background of the person in the area AR, the color difference flatness determination value becomes a large value. Therefore, as shown in FIG. 4A, a histogram can be obtained in which a large number of small values are included and large values are included little by little.

As a method for determining the representative value, as shown in FIG. 5A, the first method uses the average value of the color difference flatness determination values of each determination pixel block set in the determination area as the representative value. Further, in the second method, as shown in FIG. 5A, the color difference flatness determination value of each determination pixel block set in the determination area is made into a histogram, and the peak value is used as a representative value.

Further, in the third method, the representative value of the color difference flatness determination value of each determination pixel block set in the determination area is determined based on the average value and the standard deviation. Specifically, as shown in FIG. 5B, the standard deviation ± σ from the average value is obtained for the color difference flatness determination value of each determination pixel block. Then, the range of the standard deviation ± σ is set as the representative value calculation area, and the average value of the color difference flatness determination values belonging to the representative value calculation area is set as the representative value. If the standard deviation value is larger than the average value, the representative value is calculated for values up to twice the average value.

Note that FIG. 5B shows an example of a histogram when the object is a fake, and the standard deviation ± σ from the average value of the color difference flatness determination values is within the positive region on the horizontal axis. On the other hand, FIG. 5C shows an example of a histogram when the object is genuine. In this case, the color difference flatness determination value is concentrated on a small value, and the average value is a small value, so that the standard deviation −σ on the negative side extends to the negative region on the horizontal axis. In such a case, as shown in FIG. 5C, the representative value calculation area is determined excluding the area where the standard deviation is negative, and the average value of the color difference flatness determination values in the area is used as the representative value. And.

FIG. 4B shows an example of the distribution of representative values for a plurality of samples including genuine and fake. Sample X in the figure is genuine, but since the imaged data includes a region with a high frequency component such as the boundary between the person and the background, the representative value of the color difference component V is a little high, about "4". ing. In this way, even if it is genuine, the representative value may be slightly larger, but as a whole, the genuine representative value is concentrated on the smaller value. Therefore, by performing the authenticity determination based on the obtained representative value, it is possible to correctly perform the authenticity determination even when the determination area is not the area of a single color of the object.

[Mask processing of character area]
In a driver's license, which is an example of an identification card, the expiration date is a single color of green, blue, or gold, and a character string indicating the expiration date is printed on that part. Due to the structure of a sensor such as a scanner, color noise is likely to occur in the character string portion during reading. Therefore, it is preferable to mask the character portion and perform the color difference flatness determination on the background color portion.

FIG. 6A shows an example of masking the expiration date of the driver's license. A determination pixel block of m × n pixels (for example, 4 × 4 pixels) is set for the expiration date portion, and a white or black portion corresponding to a character is extracted. For example, a threshold value is set so that a pixel having a brightness value of 220 or more is determined to be a white pixel and a pixel having a brightness value of 50 or less is determined to be a black pixel, and white or black pixels are detected. When white or black pixels (referred to as “black and white pixels”) are detected, the area around the black and white pixels is set as a mask area as illustrated in FIG. 6 (B). In this way, black-and-white pixels are detected for all the pixels, and the periphery of the black-and-white pixels is masked. FIG. 6A shows an example after mask processing. When the mask processing is completed for all the areas, the above-mentioned color difference flatness determination value is obtained for the area excluding the mask area, and the authenticity is determined. This makes it possible to prevent a decrease in determination accuracy due to the character portion.

[Method of authenticity judgment]
The authenticity determination unit 13 determines whether the object is genuine or fake based on the color difference flatness determination value or the representative value. As a method of authenticity determination, the simplest method compares the obtained color difference flatness determination value or representative value with a predetermined threshold value, and if it is smaller than the threshold value, determines that the object is genuine.

Instead, the Mahalanobis analysis may be performed to determine the authenticity using the obtained color difference flatness determination value or representative value. Specifically, the color difference flatness determination value of the imaged data obtained from the scanner 2 is used as a calculation element of the so-called Mahalanobis distance (that is, a kind of scale indicating the degree of separation between the standard genuine and fake). The authenticity is determined by comparing the calculated Mahalanobis distance with the preset authenticity threshold. In addition, in order to calculate the Mahalanobis distance, it is necessary to learn by using a certain number of real ID sample samples (creating parameters for calculating the distance). As a specific determination method, for example, if the authenticity threshold is set to "4" and the Mahalanobis distance is equal to or greater than the authenticity threshold, the object C corresponding to the imaging data is determined to be "fake". On the other hand, if the Mahalanobis distance is less than the authenticity threshold, the identification card C corresponding to the imaged data is determined to be "genuine". References that can be referred to when setting the authenticity threshold include, for example,
Reference "Proposal and comparison of threshold setting method in MT method, Masanari Abe, Shinichi Matsuda"("http://www.seto.nanzan-u.ac.jp/msie/nas/academia/vol_013pdf/13-023" -033.pdf "etc.)
And so on.

[Determination process]
Next, the entire flow of the determination process by the determination device 100 will be described. FIG. 7 is a flowchart of the determination process. This process is mainly carried out by the processing unit 1 of the determination device 100 executing a program prepared in advance.

First, when the user sets the identification card C in the scanner 2 and operates the operation unit 4 to give a reading instruction, the processing unit 1 detects this and gives an imaging instruction of the identification card C to the scanner 2 ( Step S11). The scanner 2 images the identification card C based on the image pickup instruction, and transmits the image pickup data to the processing unit 1. When the processing unit 1 receives the imaging data (step S12: Yes), the processing unit 1 cuts out the image of the determination region from the imaging data and performs YUV conversion to generate the luminance component Y and the color difference components U and V (step S13). When the determination area includes the character area as described above, the processing unit 1 performs mask processing and then YUV conversion.

Next, the processing unit 1 performs a color difference flatness determination based on the obtained color difference components U and V to calculate a color difference flatness determination value (step S14), and further based on the obtained color difference flatness determination value. The authenticity of the object is determined as to whether it is genuine or fake (step S15). When a determination region other than a single color is used as described above, a representative value is determined from the obtained average degree determination value, and authenticity determination is performed based on the representative value. Then, the processing unit 1 outputs the result of the authenticity determination to the display unit 15 and displays it (step S16). In this way, the determination process ends.

[Modification example]
(Modification example 1)
In the above determination process, a color-printed area is used as the determination area. In the present embodiment, since the determination is performed using the colored region such as ink, the determination is not performed on the non-colored region to which the ink or the like is not applied, that is, the white background (base material surface). Specifically, when the determination area is a portion of a photograph and the background of the photograph is white, the white area is detected using a predetermined threshold value, and the determination is not performed for that area. Further, it is preferable that the dark black portion is not included in the determination because the difference in color difference is unlikely to appear. Specifically, a threshold value is set for the black region as well, and the determination is not performed for the region darker than a certain level. As a result, it is possible to prevent a decrease in determination accuracy.

(Modification 2)
In the above embodiment, as an example of separating the imaging data into a brightness component and a color component, the RGB data of the imaging data is YUV-converted to generate the brightness component Y and the color difference components U and V. The application is not limited to this. For example, RGB data may be converted into a luminance component Y and color difference components Cb and Cr. Further, in the above embodiment, the color difference is used as the color component of the imaging data, but a parameter indicating another color component may be used. That is, parameters other than the color difference may be extracted from the imaging data as parameters indicating the color components, and the flatness of the color components may be obtained based on the parameters. The "color flatness" in the present invention is a concept that includes not only the color difference flatness but also the flatness calculated based on the parameters indicating the color components other than the color difference as described above.

(Modification example 3)
In the above embodiment, the scanner 2 reads the type data stored in the identification card C to specify the determination area, and reads the identification card C in the determination area to generate the imaging data. Instead, the scanner 2 images the entire surface of the identification card C to create the imaging data of the entire surface, and the processing unit 11 identifies the type of the identification card C and the determination area by image recognition of the imaged data on the entire surface. You may cut out only the necessary part as imaging data. In this method, it is not necessary to read the type data from the identification card C, so that a reader or the like attached to the scanner 2 is not required, and the cost of the system can be reduced.

(Modification example 4)
In the above embodiment, the representative value is determined from the color difference flatness determination values of the plurality of determination pixel blocks only when the determination area other than the single color is used, and the authenticity determination is performed using the representative values. Even when the determination area of a single color is used, the representative value may be determined from the color difference flatness determination values of the plurality of determination pixel blocks, and the authenticity determination may be performed using the representative values. As a result, the influence of scratches, stains, etc. on the face of the identification card C can be reduced, and a more accurate determination can be made.

1 Processing unit 2 Scanner 3 Storage unit 4 Operation unit 5 Display unit 11 Conversion unit 12 Flatness judgment unit 13 Authenticity judgment unit 100 Judgment device

Claims (8)

An acquisition means for acquiring captured data of a part of an object, and
A color component generating means for generating a color component of the captured image from the captured data,
A color flatness calculating means for determining the color flatness of the captured image based on the color component, and
Based on the determined color flatness, a printing determination means for determining whether or not a part of the object is printed by the area gradation method, and
A mask processing means for performing mask processing on a character area included in the imaged data is provided.
The mask processing means detects a pixel having a brightness component of the first predetermined value or more and a pixel having a brightness component of the second predetermined value or less among a plurality of pixels included in the captured image, and the pixel and the periphery of the pixel. The pixel area is masked as the character area,
The determination device, wherein the color component generating means generates the color component in a region other than the character region masked by the masking process. The color flatness calculating means is characterized in that the color flatness is calculated based on any one of the average deviation, the dispersion and the standard deviation of the color components in a plurality of pixel regions constituting the captured image. Judgment device described in. The color flatness calculating means is characterized in that it performs frequency analysis of color components in a plurality of pixel regions constituting the captured image and calculates the color flatness based on the magnitude of frequency components other than the DC component. The determination device according to claim 1. The color flatness calculating means calculates the color flatness for each of a plurality of pixel regions constituting the captured image, determines a representative value of the obtained plurality of color flatnesses, and determines the representative values of the obtained plurality of color flatnesses.
Any one of claims 1 to 3, wherein the print determination means determines whether or not a part of the object is printed in an area gradation system based on the representative value. Judgment device according to. The color flatness calculating means is within a range of a standard deviation based on the average value of the plurality of color flatnesses, the peak value in the histogram of the plurality of color flatnesses, and the average value of the plurality of color flatnesses. The determination device according to claim 4, wherein any one of the average values of the color flatness in the above is used as the representative value. The imaging data is RGB data, and is
The determination device according to any one of claims 1 to 5 , wherein the color component generating means generates a color difference component from the RGB data to obtain the color component. Any one of claims 1 to 6 , wherein when a part of the object is printed in an area gradation system, the object is provided with an authenticity determination means for determining that the object is a fake. Judgment device according to. A program executed by a judgment device equipped with a computer.
Acquisition means for acquiring captured data of a part of the captured image of the object,
A color component generating means for generating a color component of a captured image from the captured data,
A color flatness calculating means for determining the color flatness of a captured image based on the color component,
A printing determination means for determining whether or not a part of the object is printed in an area gradation system based on the determined color flatness.
A mask processing means that performs mask processing on a character area included in the imaged data.
To make the computer function as
The mask processing means detects a pixel having a brightness component of the first predetermined value or more and a pixel having a brightness component of the second predetermined value or less among a plurality of pixels included in the captured image, and the pixel and the periphery of the pixel. The pixel area is masked as the character area,
The color component generating means is a program characterized in that the color component is generated in a region other than the character region masked by the masking process.

Images