JP7803176B2 - Authenticity determination device and program - Google Patents

Description

The present invention relates to an authenticity determination device and a program therefor.

When opening an account at a financial institution, you must verify your identity with an ID card (personal identification) such as a My Number card, but there has been an increase in cases where ID cards are forged or altered in order to open accounts under the guise of another person for criminal purposes such as special fraud.

At typical financial institutions, users go to the institution and submit their ID card, and the financial institution then uses the submitted ID card to verify the user's identity in person. In this case, the ID card can be carefully observed, and any anomalies can be detected by the feel of the card.

Patent documents 1 and 2 also describe authenticity determination using concentric circular patterns formed on printed materials. If this is applied to ID cards, it would be possible to determine the authenticity of submitted ID cards by reading them with a reading device such as a scanner.

Patent No. 5585915 Patent No. 5799425

Recently, online banks and other online financial institutions have become common. Some of these financial institutions verify identity by taking a photo of the face of your ID card with a camera on a smartphone or other device and sending the image to the financial institution.

In this case, identity verification is performed using only an image, rather than face-to-face, so there is a need for a way to detect counterfeit or altered ID cards. However, it is not easy to detect a fake ID card from an image alone, which has been an issue when applying online, etc.

For example, even when authenticity determination is based on concentric circular patterns as in Patent Documents 1 and 2, when photographing the pattern with a smartphone or other camera, there is room for improvement in the accuracy of the determination, in terms of the degree of reproducibility of the pattern in the photographed image.

The present invention was made in consideration of the above problems, and aims to provide an authenticity determination device that can effectively determine the authenticity of media such as identification cards.

A first invention for solving the above-mentioned problems is an apparatus for determining the authenticity of a medium on which a plurality of predetermined concentric circular colored patterns are formed, the apparatus having an authenticity determination means for obtaining a plurality of inspection images showing each of the predetermined concentric circular colored patterns from a photographed image of the medium, determining whether or not the predetermined concentric circular colored pattern is present on the medium based on the plurality of inspection images, and determining that the medium is genuine if it is determined that the predetermined concentric circular colored pattern is present, the authenticity determination means performing the determination using a plurality of Fourier transform images created by performing a Fourier transform on each of the inspection images, and the authenticity determination means performing pixel determination based on the plurality of Fourier transform images. an integrated image is created in which the pixel value of each pixel represents the strength of the spatial frequency, the distance from the image center to the pixel represents the height of the spatial frequency, and the direction of the line segment connecting the image center and the pixel represents the wave direction, and a discrimination value used for the discrimination is calculated from the integrated image, and the authenticity determination means, when calculating the discrimination value, calculates an average value of pixel values of pixels located at each distance from the image center of the integrated image, and creates an equation showing a tendency for the pixel value to decrease according to the distance, calculates a difference between the average value and the value of the equation for each distance, and calculates the discrimination value based on the difference . The medium is, for example, an identification card, but is not limited to this.

The present invention determines the authenticity of a medium such as an identification card based on the presence or absence of a predetermined concentric circular pattern. Specifically, the predetermined concentric circular pattern formed on the medium is deformed or lost due to deterioration when the medium is reproduced by photocopying or other methods. Therefore, if the medium is altered by pasting a copy or a printout of an edited image of the medium onto the medium, the predetermined concentric circular pattern will not appear. Therefore, the present invention can effectively detect fake media when the predetermined concentric circular pattern disappears as a result of forgery or alteration of the medium. Furthermore, the present invention makes it possible to improve the accuracy of the determination by making a comprehensive determination based on multiple test images showing multiple concentric circular patterns.

In addition, in the present invention, the authenticity determining means makes the determination using a plurality of Fourier transform images created by performing a Fourier transform on each of the inspection images.
In the present invention, the presence or absence of the specified concentric circular pattern can be determined by analyzing the Fourier transform image to determine whether the medium has a spatial frequency corresponding to the spacing between the inner and outer circles in the specified concentric circular pattern.

Furthermore, in the present invention, the authenticity determination means creates an integrated image based on the plurality of Fourier transform images, in which the pixel value of a pixel represents the strength of the spatial frequency, the distance from the center of the image to the pixel represents the height of the spatial frequency, and the direction of the line segment connecting the center of the image to the pixel represents the direction of the wave, and calculates a discrimination value to be used for the discrimination from the integrated image .
In the present invention, by using an integrated image obtained by integrating multiple Fourier transform images to determine whether or not a specified concentric circular pattern is present, comprehensive authenticity determination can be performed, thereby improving the accuracy of authenticity determination.

Furthermore, in the present invention, when calculating the discrimination value, the authenticity determination means calculates the average value of pixel values of pixels located at a distance from the image center of the integrated image for each distance from the image center, creates an equation showing the tendency of the pixel values to decrease according to the distance, calculates the difference between the average value and the value of the equation for each distance, and calculates the discrimination value based on the difference .
Whether a medium has a spatial frequency corresponding to the spacing between the inner and outer circles in the predetermined concentric pattern can be determined from the average value described above, but Fourier transform images and integrated images based on them generally have a tendency for pixel values to decrease from the center to the periphery. Therefore, in the present invention, by calculating the difference between the average value and the value of an equation that indicates the decreasing tendency of pixel values, the influence of this decreasing tendency can be avoided and the accuracy of authenticity determination can be improved.

It is also preferable that the authenticity determining means masks a predetermined area of the integrated image and excludes pixels within that area from the objects used to calculate the average value.
If a background pattern is formed over the concentric circular pattern, the influence of the background pattern may appear in the pixel values of a predetermined area of the Fourier transform image and the integrated image based on it. In this case, the accuracy of authenticity determination can be improved by masking the area and excluding it from the target for calculating the average value.

The authenticity determination device preferably includes a receiving means for receiving a photographed image of the medium from a user terminal.
This makes it possible to build, via a network, an authenticity determination system that determines the authenticity of a medium using a photographed image of the medium sent from a user terminal.

A second invention is an apparatus for determining the authenticity of a medium on which a plurality of predetermined concentric circular patterns are formed, the apparatus comprising: an authenticity determination means for obtaining a plurality of inspection images showing the individual concentric circular patterns from a photographed image of the medium, determining whether or not the predetermined concentric circular pattern is present on the medium based on the plurality of inspection images, and determining that the medium is genuine when it is determined that the predetermined concentric circular pattern is present; the authenticity determination means performs the determination using a plurality of Fourier transform images created by performing a Fourier transform on each of the inspection images, and the authenticity determination means spatially converts pixel values of pixels based on the plurality of Fourier transform images. The program creates an integrated image in which the intensity of the frequency, the distance from the center of the image to the pixel in question, is the height of the spatial frequency, and the direction of the line segment connecting the center of the image to the pixel in question, is the direction of the wave, and calculates a discrimination value to be used for the discrimination from the integrated image. When calculating the discrimination value, the authenticity determination means calculates, for each distance from the center of the image of the integrated image, the average value of the pixel values located at that distance from the center of the image, creates an equation that shows the tendency of the pixel values to decrease according to the distance, calculates the difference between the average value and the value of the equation for each distance, and calculates the discrimination value based on that difference, to function as an authenticity determination device.

This invention makes it possible to provide an authenticity determination device that can effectively determine the authenticity of media such as personal identification cards.

FIG. 1 is a diagram showing an authenticity determination system 1. FIG. 2 is a diagram showing the hardware configuration of the authenticity determination device 3. FIG. 2 is a diagram showing the hardware configuration of a user terminal 5. FIG. 2 is a block diagram showing the functions of the authenticity determination device 3. FIG. 1 shows an ID card 10. 1 is a flowchart showing an outline of the processing of the authenticity determination system 1. An example of the ID card 10 photography screen. 10 is a flowchart showing an authenticity determination method. FIG. 1 shows an image 100. 1A and 1B are diagrams illustrating normalization of an image 100. 3A to 3C are diagrams for explaining the cutting out of inspection images 200a to 200c. 3A to 3C are diagrams illustrating Fourier transform images 300a to 300c. 4A to 4C are diagrams illustrating the creation of an integrated image 400. FIG. 10 is a diagram illustrating calculation of a discriminant value R. FIG. 10 is a diagram illustrating calculation of a discriminant value R.

The following describes in detail a preferred embodiment of the present invention with reference to the drawings.

(1. Authenticity Determination System 1)
1 is a diagram showing an authenticity determination system 1 having an authenticity determination device 3 according to an embodiment of the present invention. The authenticity determination system 1 is configured by connecting the authenticity determination device 3 and a user terminal 5 via a network so that they can communicate with each other. In the authenticity determination system 1, a captured image of an ID card 10 is sent from the user terminal 5 to the authenticity determination device 3, and the authenticity determination device 3 determines the authenticity of the ID card 10 from the captured image.

Figure 2 shows the hardware configuration of the authenticity determination device 3. The authenticity determination device 3 can be implemented by a computer configured by connecting a control unit 31, a memory unit 32, a communication unit 33, etc. via a bus or the like. However, this is not limited to this and various other configurations can be used as appropriate, and the authenticity determination device 3 can also be implemented by multiple computers.

The control unit 31 is composed of a CPU, ROM, RAM, etc. The CPU loads programs related to the processing of the authenticity determination device 3 stored in storage media such as the memory unit 32 and ROM into a work area on the RAM and executes them. The ROM is non-volatile memory and permanently stores programs such as the boot program and BIOS, as well as data. The RAM is volatile memory and temporarily stores programs and data loaded from the memory unit 32, ROM, etc., and also has a work area used by the control unit 31 to perform various processes.

The memory unit 32 is a hard disk drive, solid state drive, flash memory, etc., and stores the programs executed by the authenticity determination device 3 during the processing described below, data required for program execution, the OS, etc.

The communication unit 33 is a communication interface that mediates communication via a network and communicates with the user terminal 5.

The user terminal 5 is a terminal carried by the user, which takes a picture of the ID card 10 and transmits the captured image to the authenticity determination device 3. The user terminal 5 may be, for example, a mobile terminal such as a smartphone or tablet.

Figure 3 is a diagram showing the hardware configuration of the user terminal 5. As shown in Figure 3, the user terminal 5 is configured by connecting a control unit 51, a memory unit 52, a display unit 53, an input unit 54, a communication unit 55, a camera 56, an audio input/output unit 57, etc. via a bus or the like. However, this is not limited to this, and various other configurations can be used as appropriate.

The control unit 51, memory unit 52, and communication unit 55 are substantially the same as the control unit 31, memory unit 32, and communication unit 33 described above. The display unit 53 has a display device such as an LCD panel, and a touch panel is provided as the input unit 54 for inputting data to the user terminal 5. The audio input/output unit 57 includes a microphone and speaker used for audio input and output.

Camera 56 is an area camera composed of an optical lens, an imaging element such as a CCD (Charge Coupled Device) or CMOS (Complementary Metal-Oxide Semiconductor), an A/D (Analog/Digital) converter, etc. Camera 56 photoelectrically converts the subject image input through the optical lens using the imaging element to generate an analog image signal. The A/D converter then converts the analog image signal into digital image data.

Figure 4 is a block diagram showing the functions of the authenticity determination device 3. The authenticity determination device 3 has a receiving means 301, an authenticity determination means 302, etc.

The receiving means 301 allows the control unit 31 of the authenticity determination device 3 to receive a captured image of the ID card 10 from the user terminal 5 via the communication unit 33.

The authenticity determination means 302 determines the authenticity of the ID card 10 using a photographed image of the ID card 10 received by the control unit 31 of the authenticity determination device 3 from the user terminal 5. Details of authenticity determination will be described later.

(2. ID Card 10)
In this embodiment, the ID card 10 is an identification card that is the subject of authentication determination. An identification card is a medium used for identity verification by financial institutions, mobile phone carriers, various administrative agencies, etc., and in this embodiment, the ID card 10 is a My Number card.

Figure 5(a) is a diagram showing an outline of the ID card 10 (My Number Card), with some of the face of the ID card 10 omitted.

Various information is printed on the face of the ID card 10. For example, in the case of a My Number card, the user's face image, name, address, date of birth, expiration date, etc. are printed as face information.

Figure 5(b) is an enlarged view of a portion of the background area of the face information of ID card 10. As shown in Figure 5(b), a plurality of predetermined concentric colored patterns (hereinafter referred to as concentric colored patterns) are formed in a predetermined portion of the background area. In each concentric colored pattern, multiple circles are arranged at equal intervals in the inner and outer directions. This interval, i.e., the difference in radius between the inner and outer circles, is set to a predetermined value.

(3. Overview of the processing of the authenticity determination system 1)
Fig. 6 is a flowchart showing an outline of the processing executed by the authenticity determination system 1. S1 to S2 and S7 to S8 in Fig. 6 are processing executed by the control unit 51 of the user terminal 5 by controlling each unit of the user terminal 5, and S3 to S6 are processing executed by the control unit 31 of the authenticity determination device 3 by controlling each unit of the authenticity determination device 3.

In this embodiment, the user first operates the user terminal 5 to launch a dedicated application program pre-stored in the memory unit 52. The user terminal 5 then displays a menu screen (not shown) on the display unit 53, activates the camera 56 in accordance with the user's selection on the menu screen, and photographs the face of the ID card 10 (S1).

Figure 7 shows an example of the ID card 10 capture screen at this time. In this embodiment, when the camera is started, instructions are displayed on the display unit 53 of the user terminal 5 to prompt the user to place the ID card 10 within a specified frame 531 before capturing the ID card 10.

The user terminal 5 transmits a photographed image of the ID card 10 to the authenticity determination device 3 (S2). This image is a color image (RGB image) in which the pixel colors are determined by the gradation values of each of the R (red), G (green), and B (blue) components.

The authenticity determination device 3 receives the photographed image of the ID card 10 (S3) and determines the authenticity of the ID card 10 from the photographed image of the ID card 10 (S4). In this embodiment, the authenticity determination is performed based on a predetermined concentric circular pattern formed on the ID card 10. Details of the authenticity determination method based on the concentric circular pattern will be described later.

If the authenticity determination device 3 determines that the ID card 10 is fake (S5; NO), it transmits a message to the user terminal 5 indicating that the ID card 10 is fake (S6). When the user terminal 5 receives a message indicating that the ID card 10 is fake (S7), it displays this message on the display unit 53 (S8) and ends the process.

On the other hand, if the authenticity determination device 3 determines that the ID card 10 is authentic (S5; YES), it completes the authenticity determination process. For example, when authenticity determination is performed as identity verification when opening an account at an online financial institution, after the authenticity determination process, user information (such as name and date of birth) on the face of the ID card 10 can be obtained from a photographed image of the ID card 10 as information required to open the account, and this information can be registered, and the user information can be stored in the memory unit 52.

(4. Authentication based on concentric circular patterns)
As described above, a plurality of predetermined concentric colored patterns are formed on the face of the ID card 10. These concentric colored patterns deteriorate and become distorted or disappear when the ID card 10 is reproduced by photocopying or other methods. Therefore, if the ID card 10 is falsified by attaching a copy of the face of the ID card 10 or a printout of an edited image of the face of the ID card 10 to the face of the ID card 10, the predetermined concentric colored patterns will not appear. Therefore, in this embodiment, a counterfeit ID card 10 is detected based on the presence or absence of such predetermined concentric colored patterns. Note that the methods of forgery and falsification of counterfeit ID cards 10 that can be detected in this embodiment are not limited to those described above, and any method that prevents the predetermined concentric colored patterns formed on the ID card 10 from appearing may be used.

Figure 8 is a flowchart showing the steps of the authenticity determination method in S4 of Figure 6.

In S4, the authenticity determination device 3 first extracts an area containing multiple concentric circular patterns from the captured image of the ID card 10 (S401). This area is set in advance to match the position of the concentric circular patterns. Reference numeral 11 in Figure 5(b) is an example of this area, and Figure 9 is an example of image 100 obtained by extracting this area 11. Image 100 contains three concentric circular patterns. Note that in this embodiment, an area containing multiple concentric circular patterns is extracted from one location in the captured image, but areas containing concentric circular patterns may also be extracted from multiple locations in the captured image, reducing the effects of partial blurring of the captured image. In this case, the processes of S402 to S406, described below, are performed for each extracted image.

The authenticity determination device 3 then converts the image 100 to a grayscale image (S402). In this embodiment, grayscale conversion is performed in which the gradation value of the G component of each pixel in the image 100 is used as the gradation value of each pixel in the grayscale image. This is because the periphery of the concentric circular pattern in the actual photographed image of the ID card 10 is generally reddish. To avoid this effect, the gradation value of the G component, which is the complementary color of red, is used as the gradation value of the grayscale image. However, the method of creating a grayscale image is not limited to this. For example, as in normal grayscale conversion, the weighted sum of the gradation values of the R, G, and B components of each pixel in the image 100 may be used as the gradation value of each pixel in the grayscale image. In some cases, the gradation value of the R or B component of each pixel in the image 100 may also be used as the gradation value of each pixel in the grayscale image.

Next, the authenticity determination device 3 performs edge enhancement processing (S403) on the grayscale converted image 100. Edge enhancement can be performed by image processing using a known filter, and by appropriately determining the coefficient values of the filter matrix, the edges in the image 100 are enhanced.

The authenticity determination device 3 then performs a normalization process on the image 100 after edge enhancement (S404). The normalization process in this embodiment converts (expands) the range of tone values in the image 100 to a predetermined range, thereby increasing the light-dark contrast of the image 100.

More specifically, in the histogram of gradation values for each pixel in image 100 shown in Figure 10, the lower a (%) value is taken as the minimum value MINn and the upper b (%) value is taken as the maximum value MAXn, and the gradation value range from MINn to MAXn is converted to a range from normLowVal (normalized minimum value) to normHighVal (normalized maximum value). For example, the values of a and b are 10 (%), and the values of normLowVal and normHighVal are 50 and 200, respectively, but this is not limited to these values.

In S404, the gradation value Pn of each pixel of the image 100 is converted using the following formula (1) to obtain the gradation value NPn after normalization processing. If the value of the gradation value NPn calculated using formula (1) is 0 or less, the gradation value NPn is set to 0, and if the value of the gradation value NPn calculated using formula (1) is 255 or more, the gradation value NPn is set to 255.
NPn=normLowVal+(normHighVal−normLowVal)×((Pn−MINn)/(MAXn−MINn))…(1)

The authenticity determination device 3 converts the normalized image 100 to a predetermined resolution (S405). Resolution conversion can be performed using a known method such as the bilinear method.

The authenticity determination device 3 then extracts the ranges of each concentric circular pattern from the resolution-converted image 100 to create an inspection image (S406). In this embodiment, image 100 contains three concentric circular patterns, and in S406, three ranges r1 to r3 corresponding to each concentric circular pattern are extracted, as shown in Figure 11. This results in multiple inspection images 200a to 200c showing each concentric circular pattern. The vertical and horizontal sizes of ranges r1 to r3 are the same, and ranges r1 and r2, and ranges r2 and r3 partially overlap horizontally in image 100.

The authenticity determination device 3 performs a Fourier transform on each of the inspection images 200a-200c to create multiple Fourier transform images (S407). Figure 12(a) shows an example of Fourier transform images 300a-300c created by the processing up to S407 from an actual image of the ID card 10.

Fourier transform images 300a-300c are images that show the spatial frequencies appearing in inspection images 200a-200c using the pixel values of each pixel. As shown in Figure 12(b), the distance d from the image center c to pixel P corresponds to the height of the spatial frequency, and the direction (directional angle) α of the line segment connecting the image center c and pixel P corresponds to the direction of the wave.

The pixel value of pixel P represents the strength of the spatial frequency (power spectrum) as a value between 0 and 1. The Fourier transform images 300a to 300c in Figure 12(a) are grayscale images in which pixel colors are represented as white when the spatial frequency power spectrum is large (pixel value 1) and black when the power spectrum is small (pixel value 0).

In the Fourier transform images 300a to 300c in Figure 12(a), the areas near the horizontal axis (x-axis) and vertical axis (y-axis) passing through the image center c appear white. This is due to the influence of a background pattern (not shown) formed on the ID card 10 along with the concentric circular pattern.

The authenticity determination device 3 performs low-pass filtering on the Fourier transform images 300a-300c (S408). Low-pass filtering is a known process, but in this embodiment, each pixel in the Fourier transform images 300a-300c is treated as a pixel of interest, and the average of the seven highest pixel values among the pixel values of the pixel of interest and its eight surrounding pixels is used to determine the pixel value of the pixel of interest. This reduces the influence of noise that appears as dark areas in the image. Note that there are no particular restrictions on how many highest pixel values are averaged in the above process.

Then, the authenticity determination device 3 merges the three Fourier transform images 300a to 300c to create a merged image (S409).

In S409, the authenticity determination device 3 first calculates the average pixel value DM for pixels within a predetermined region of the Fourier transform image 300a, and then calculates the difference by subtracting the average pixel value DM from the pixel value of each pixel of the Fourier transform image 300a. Symbol D in Figure 13(a) shows an example of the above-mentioned predetermined region on the Fourier transform image 300a in Figure 12(a). The above-mentioned difference takes on values between -1 and 1.

This process is also performed on the other Fourier transform images 300b and 300c, and the above difference is calculated for each pixel of the Fourier transform images 300a-300c. Because the Fourier transform images 300a-300c may have differences such as being generally bright or generally dark, the overall brightness and darkness of the Fourier transform images 300a-300c is averaged out by taking the difference from the average value DM.

Then, as shown in Figure 13(b), the authenticity determination device 3 determines the pixel value Pf of each pixel P in the integrated image 400 as the average of the top two differences ΔPa, ΔPb, and ΔPc found for pixels P at the same position in the Fourier transform images 300a to 300c. By performing the above process for each pixel P in the integrated image 400, in S409 an integrated image 400 of the same size as the Fourier transform images 300a to 300c is obtained.

Like Fourier transform images 300a-300c, integrated image 400 is an image in which the pixel value Pf of pixel P represents the strength of the spatial frequency, the distance from the image center to pixel P represents the height of the spatial frequency, and the direction of the line segment connecting the image center to pixel P represents the direction of the wave. Furthermore, in integrated image 400, the pixel value Pf of pixel P at one position may be the average of ΔPa and ΔPb, while the pixel value Pf of pixel P at another position may be the average of ΔPb and ΔPc, resulting in a stronger emphasis on the bright areas of each Fourier transform image 300a-300c. Furthermore, the pixel value Pf of pixel P in integrated image 400 can be the maximum value of the differences ΔPa, ΔPb, and ΔPc, or it can also be the average value of the differences ΔPa, ΔPb, and ΔPc.

The authenticity determination device 3 then performs masking of a predetermined region including the x-axis and y-axis of the integrated image 400 (S410). That is, as mentioned above, in the Fourier transform images 300a to 300c in Figure 12(a), pixel values are large near the x-axis and y-axis due to the influence of a background pattern (not shown), and this is also true for the integrated image 400 based on the Fourier transform images 300a to 300c. The pixel values Pf near the x-axis and y-axis of the integrated image 400 cause variation in the discrimination value (described below), which can reduce the accuracy of authenticity determination.

For this reason, in S410, a specific area including the x-axis and y-axis of the integrated image 400 is masked to prevent a decrease in the accuracy of the authenticity determination. The masking process is performed by setting the pixel values Pf of all pixels P within the above area to a specific value. The specific value can be the average of the pixel values Pf of pixels P in other areas (for example, areas near the center of the image).

The authenticity determination device 3 performs a normalization process on the integrated image 400 after the masking process (S411). This normalization process, like the above, converts the range of pixel values Pf of the integrated image 400. That is, the range of pixel values Pf of the integrated image 400 is -1 to 1, but the normalization process of S411 converts the range of pixel values Pf of the integrated image 400 to 0 to 1. More specifically, assuming that the pixel value of each pixel P of the integrated image 400 is Pf, and the maximum and minimum pixel values Pf in the integrated image 400 are Pfmax and Pfmin, respectively, the pixel value NPf after the normalization process of each pixel P of the integrated image 400 is calculated by the following equation (2):
NPf＝(Pf－Pfmin)／(Pfmax－Pfmin)...(2)

Figure 14(a) shows an example of an integrated image 400 obtained by the processing up to S411. The white stripes extending vertically and horizontally in the integrated image 400 are areas that were masked in S410. The authenticity determination device 3 calculates a discriminant value used for authenticity determination from this integrated image 400 (S412).

In this embodiment, in S412, as shown in FIG. 14(b), the average value PM of the pixel values NPf of pixels P located at positions M that are a distance d away from the image center c of the integrated image 400 is calculated for each distance d. These pixels P are arranged circumferentially, and in S412, the average value PM is calculated for pixels P in the first and second quadrants of the integrated image 400 (the upper half of the integrated image 400). This is because the Fourier transform images 300a-300c and the integrated image 400 based on them are point-symmetric images with respect to the image center c.

In addition, the area that has undergone the masking process is excluded from the calculation of the average value PM. Note that since the concentric circular pattern does not have directional shape characteristics, the average value PM may be calculated for only the first or second quadrant. However, calculating the average value PM for the pixels in both the first and second quadrants improves the accuracy of authenticity determination.

Figure 15(a) is a graph in which the horizontal axis represents distance d (= dmin, dmin+1, dmin+2, ..., dmax-2, dmax-1, dmax) from the image center c, and the vertical axis represents PM, with the average PM value PM(d) shown as a solid line for each distance d. PM(d) can only be calculated for distance d within the target range. Distance d is a natural number, and the target range can be defined as appropriate.

In this embodiment, whether the ID card 10 has the above-mentioned specified concentric circular pattern can be determined by the value of PM(d). In other words, if the value of PM(d) is high at the distance d, which corresponds to the distance (spatial frequency) between the inner and outer circles in the specified concentric circular pattern, it can be determined that the ID card 10 has the above-mentioned specified concentric circular pattern.

However, in Fourier transform images 300a-300c, there is a general tendency for pixel values to decrease as you move from the image center c to the periphery, and a similar decreasing trend also appears in the pixel values NPf of integrated image 400. In this embodiment, processing is performed to prevent this tendency from affecting the judgment, thereby improving the accuracy of authenticity judgment.

That is, the authenticity determination device 3 first calculates the average dmin+1 of multiple (three in this embodiment) distances dmin, dmin+1, and dmin+2 near the lower limit dmin of the target range, and the average avePMmin of PM(d) calculated for these multiple distances dmin, dmin+1, and dmin+2. It also calculates the average dmax-1 of multiple (three in this embodiment) distances dmax-2, dmax-1, and dmax near the upper limit dmax of the target range, and the average avePMmax of PM(d) calculated for these multiple distances dmax-2, dmax-1, and dmax.

Then, as shown in Figure 15(a), the equation L(d) for the line connecting the two points (dmin+1, avePMmin) and (dmax-1, avePMmax) is created as an equation that indicates the decreasing trend in pixel value NPf according to distance d, and the difference ΔR(d) is calculated by subtracting the value of L(d) from the value of PM(d) for each distance d between dmin and dmax. This prevents the decreasing trend in pixel value NPf in the integrated image 400 from affecting the judgment.

15(b) shows the value of ΔR(d) for each distance d, with the horizontal axis representing the distance d from the image center c and the vertical axis representing ΔR(d). In this embodiment, particularly, the value of ΔR(d _n ) is found for each of multiple (seven in this embodiment) interest distances d _n (=d ₁ to d ₇ ) located between d min and d max , and a weighted sum Rs is calculated using W _n (=W ₁ to W ₇ ) as weighting coefficients, as shown in the following equation (3):
Rs=W ₁ ΔR(d ₁ )+W ₂ ΔR(d ₂ )+…+W ₆ ΔR(d ₆ )+W ₇ ΔR(d ₇ )…(3)

The number and the distances of interest _dn used to calculate the weighted sum Rs can be determined as appropriate to include the distance d corresponding to the distance (spatial frequency) between the inner and outer circles in the predetermined concentric circular pattern formed on the ID card 10. The weighting coefficient _Wn is determined as appropriate so that it becomes larger the closer it is to the distance d.

The authenticity determination device 3 calculates the discrimination value R from the weighted sum Rs using the following formula (4): Here, the maximum value of R is 1000, and if Rs+0.1 is smaller than 0.01, R=1000. Note that the method for calculating the discrimination value R is not particularly limited, and the weighted sum Rs itself may be used as the discrimination value R.
R = 10/(Rs + 0.1)…(4)

The authenticity determination device 3 compares the discrimination value R with the reference value (S413). If the discrimination value R is greater than the reference value, it determines that the specified concentric circular pattern is not present on the ID card 10 (S414; NO), and determines that the ID card 10 is false (S415). On the other hand, if the discrimination value R is equal to or less than the reference value, it determines that the specified concentric circular pattern is present on the ID card 10 (S414; YES), and determines that the ID card 10 is true (S416), and the process ends.

As described above, in this embodiment, the authenticity of the ID card 10 is determined based on the presence or absence of a predetermined concentric circular color pattern. Specifically, the predetermined concentric circular color pattern formed on the ID card 10 is deformed or lost due to deterioration when the ID card 10 is reproduced by photocopying or other methods. Therefore, if the ID card 10 is falsified by attaching a copy of the face of the card or a printout of an edited image of the face of the card to the face of the ID card 10, the predetermined concentric circular color pattern will not appear. Therefore, in this embodiment, a fake ID card 10 can be effectively detected when the predetermined concentric circular color pattern disappears as a result of forgery or alteration of the ID card 10. Furthermore, in this embodiment, the accuracy of the determination can be improved by making a comprehensive determination based on multiple test images 200a-200c showing multiple concentric circular color patterns.

In particular, in this embodiment, the presence or absence of the specified concentric circular pattern can be determined by analyzing the Fourier transform images 300a-300c to determine whether the ID card 10 has a spatial frequency corresponding to the spacing between the inner and outer circles in the specified concentric circular pattern.

In addition, in this embodiment, by using the integrated image 400 obtained by integrating multiple Fourier transform images 300a-300c to determine the presence or absence of a specified concentric circular pattern, comprehensive authenticity determination can be performed, thereby improving the accuracy of authenticity determination.

Whether the ID card 10 has a spatial frequency corresponding to the spacing between the inner and outer circles in the predetermined concentric circular pattern can be determined from the value PM(d). However, in the Fourier transform images 300a-300c and the integrated image 400 derived from them, there is a general tendency for pixel values to decrease as you move from the center c of the image to the periphery. Therefore, in this embodiment, by taking the difference ΔR(d) between PM(d) and the value L(d) of the equation that indicates the decreasing trend in pixel values, the influence of this decreasing trend can be avoided and the accuracy of authenticity determination can be improved.

Furthermore, if a background pattern is formed over the concentric circular pattern, the influence of the background pattern may appear in the pixel values near the x-axis and y-axis of the Fourier transform images 300a-300c and the integrated image 400 based on these. In this case, the accuracy of authenticity determination can be improved by masking a specific area including the x-axis and y-axis and excluding it from the target for calculating the value PM(d). Note that, depending on the background pattern, the influence of the background pattern may appear in a specific area other than near the x-axis and y-axis of the integrated image 400; in such cases, the specific area can also be masked using the same method as described above.

In addition, in this embodiment, an authenticity determination system 1 that determines the authenticity of an ID card 10 from a photographed image of the ID card 10 sent from a user terminal 5 can be constructed using a network.

However, the present invention is not limited to the embodiments described above. For example, in this embodiment, authenticity determination is performed as a means of identity verification when opening an account, but the application scenarios or purposes of authenticity determination in this embodiment are not particularly limited.

In addition, in this embodiment, the authenticity determination process is performed by the authenticity determination device 3 that is communicatively connected to the user terminal 5, but the authenticity determination process can also be performed on the user terminal 5 side.

In addition, the normalization processing in S404 and S411 in Figure 8, the edge enhancement processing in S403, and the low-pass filter processing in S408 are performed as needed and can be omitted.

Although the ID card 10 is a My Number card, it is not limited to this and can be applied to any identification card as long as it has a concentric circular pattern. Furthermore, the application of the present invention is not limited to identification cards, and it can also be applied to verifying the authenticity of various media other than identification cards, such as gift certificates.

Preferred embodiments of the present invention have been described above with reference to the accompanying drawings, but the present invention is not limited to these examples. It is clear that those skilled in the art will be able to conceive of various modifications and alterations within the scope of the technical ideas disclosed in this application, and it is understood that these also naturally fall within the technical scope of the present invention.

1: Authenticity determination system 3: Authenticity determination device 5: User terminal 10: ID card 100: Images 200a, 200b, 200c: Inspection images 300a, 300b, 300c: Fourier transform image 301: Receiving means 302: Authenticity determination means 400: Integrated image

Claims (5)

An apparatus for determining authenticity of a medium on which a plurality of predetermined concentric circular patterns are formed,
an authenticity determination means for obtaining a plurality of inspection images showing each of the concentric circular patterns from a photographed image of the medium, determining whether or not the predetermined concentric circular pattern is present on the medium based on the plurality of inspection images, and determining that the predetermined concentric circular pattern is present on the medium when it is determined that the predetermined concentric circular pattern is present ;
the authenticity determining means performs the determination using a plurality of Fourier transform images created by performing a Fourier transform on each of the inspection images;
The authenticity determination means creates an integrated image based on the plurality of Fourier transform images, in which the pixel value of a pixel represents the strength of the spatial frequency, the distance from the image center to the pixel represents the height of the spatial frequency, and the direction of the line segment connecting the image center and the pixel represents the direction of the wave, and calculates a discrimination value used for the discrimination from the integrated image,
When calculating the discrimination value, the authenticity determining means
For each distance from the center of the integrated image, an average value of pixel values of pixels at a position that is a distance away from the center of the image is calculated;
creating an equation that indicates a decreasing tendency of the pixel value according to the distance;
An authenticity determination device characterized in that a difference between the average value and the value of the formula is calculated for each distance, and the discrimination value is calculated based on the difference . The authenticity determination device according to claim 1, characterized in that the medium is an identification card. 3. The authenticity determination device according to claim 1 , wherein the authenticity determination means masks a predetermined area of the integrated image and excludes pixels within the masked area from the pixels used to calculate the average value. 4. The authenticity determining device according to claim 1 , further comprising a receiving means for receiving a photographed image of the medium from a user terminal. Computer,
An apparatus for determining authenticity of a medium on which a plurality of predetermined concentric circular patterns are formed,
an authenticity determination means for obtaining a plurality of inspection images showing each of the concentric circular patterns from a photographed image of the medium, determining whether or not the predetermined concentric circular pattern is present on the medium based on the plurality of inspection images, and determining that the predetermined concentric circular pattern is present on the medium when it is determined that the predetermined concentric circular pattern is present ;
the authenticity determining means performs the determination using a plurality of Fourier transform images created by performing a Fourier transform on each of the inspection images;
The authenticity determination means creates an integrated image based on the plurality of Fourier transform images, in which the pixel value of a pixel represents the strength of the spatial frequency, the distance from the image center to the pixel represents the height of the spatial frequency, and the direction of the line segment connecting the image center and the pixel represents the direction of the wave, and calculates a discrimination value used for the discrimination from the integrated image,
When calculating the discrimination value, the authenticity determining means
For each distance from the center of the integrated image, an average value of pixel values of pixels at a position that is a distance away from the center of the image is calculated;
creating an equation that indicates a decreasing tendency of the pixel value according to the distance;
A program for causing the device to function as an authenticity determination device that calculates the difference between the average value and the value of the formula for each distance and calculates the discrimination value based on the difference .