Authenticity verification device and program
The UV-based authenticity determination device addresses the issues of surface damage and counterfeiting in ID cards by using fluorescence characteristics and threshold comparisons, ensuring accurate and robust counterfeit detection.
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- DAI NIPPON PRINTING CO LTD
- Filing Date
- 2022-12-13
- Publication Date
- 2026-07-29
AI Technical Summary
Existing methods for determining the authenticity of ID cards are compromised by surface scratches and dirt, leading to reduced accuracy and potential misidentification of legitimate cards as counterfeit, and are vulnerable to counterfeiting through alterations with false information on paper attachments.
A genuineness determination device that utilizes UV illumination to capture images of ID cards, extracts substrate and non-substrate areas based on fluorescence characteristics, and performs noise reduction and threshold comparisons to accurately determine authenticity, resistant to surface damage and alterations.
The method effectively detects counterfeit ID cards by utilizing fluorescence properties of the card's substrate, maintaining high accuracy despite surface scratches and dirt, and prevents legitimate cards from being falsely identified as fake.
Smart Images
Figure 0007896486000001 
Figure 0007896486000002 
Figure 0007896486000003
Abstract
Description
Technical Field
[0006] , , , , , , , , ,
[0005] , ,
[0001] The present invention relates to a forgery determination device and its program.
Background Art
[0002] When opening an account at a financial institution or concluding a contract with a mobile carrier, etc., it is necessary to confirm the identity using an ID card (personal identification card) such as a driver's license. However, for the purpose of crimes such as special fraud, cases of forging ID cards are increasing in order to impersonate someone else and open an account or conclude a contract with a mobile carrier.
[0003] On the other hand, for example, in a normal financial institution, it is common for a user to visit the financial institution and submit an ID card, and the financial institution side confirms the identity in person using the submitted ID card. The financial institution side can detect forgery of the ID card by visual inspection or touch.
[0004] Also, Patent Documents 1 and 2 describe a forgery determination device that automatically determines the authenticity of an ID card from an image obtained by reading the face of the ID card with a scanner or the like. It is also possible to determine the authenticity of the submitted ID card by reading it with a scanner or the like.
Prior Art Documents
Patent Documents
[0005]
Patent Document 1
Patent Document 2
Summary of the Invention
Problems to be Solved by the Invention
[0006] One challenge in determining the authenticity of an ID card from an image of its surface is that prolonged carrying of the card can cause scratches and dirt to accumulate on its surface, significantly reducing the accuracy of the authentication process. This can lead to problems such as the ID card becoming undetectable or a legitimate ID card being mistakenly identified as a counterfeit.
[0007] Furthermore, one method of counterfeiting ID cards involves attaching a piece of paper with false card information printed on it to a genuine ID card. Therefore, a method for determining the authenticity of an ID card using an image of the card's face is desired that can solve the above-mentioned problems while also being able to deal with such counterfeiting.
[0008] The present invention has been made in view of the above-mentioned problems, and aims to provide a genuineness determination device, etc., that can suitably determine the authenticity of an identity verification document. [Means for solving the problem]
[0009] The first invention for solving the aforementioned problems is a genuineness determination device for determining the authenticity of an identity document, comprising: an acquisition means for obtaining a UV inspection image and a UV substrate image, respectively, by cutting out the inspection area and the substrate area of the identity document from a UV image taken under UV (Ultra Violet) illumination of the identity document to be determined; and a genuineness determination means for determining the authenticity of the identity document to be determined by using the UV substrate image to determine the range of the substrate color of the identity document in the UV image, using that range to calculate the area of the non-substrate portion that has different fluorescence characteristics from the substrate from the UV inspection image, and comparing that area with a threshold.
[0010] In this invention, the authenticity of an identification document is determined from a UV image taken of the identification document under UV illumination, utilizing the fact that the substrate of the identification document has predetermined fluorescent properties. This makes it possible to detect counterfeit identification documents that have been altered by printing false information on a piece of paper or the like with different fluorescent properties from the substrate of the identification document and attaching it. This determination method is resistant to scratches and dirt on the substrate, and the accuracy of authenticity determination does not decrease significantly even if scratches or dirt occur on the surface of the identification document.
[0011] The authenticity determination means, in order to calculate the area, creates a substrate portion extraction image from the UV inspection image showing the substrate portion and the non-substrate portion, and it is desirable that when creating the substrate portion extraction image, noise reduction processing is performed so as not to include the text portion on the substrate as the non-substrate portion. This prevents the text on the base material of the identification document from being extracted as non-base material, thus preventing a legitimate identification document from being mistakenly identified as a fake one.
[0012] The acquisition means obtains a visible light inspection image by cutting out the inspection area from a visible light image taken of the identification document under visible light illumination, and the authenticity determination means preferably creates a character portion extraction image by extracting the character portion from the visible light inspection image in order to calculate the area, and calculates the area excluding the character portion as the area. This allows for the calculation of the area of the non-substrate portion excluding the text portion, and the calculation of the area of the non-substrate portion excluding the text on the non-substrate portion. Furthermore, even if the text on the substrate remains in the non-substrate portion of the substrate portion extraction image for some reason, the influence of that text on authenticity determination can be reduced.
[0013] The UV image is a photograph of an identification document inserted into a scanner under UV illumination, and multiple substrate regions are set. Preferably, the range of the substrate color is determined from the UV substrate image of the substrate region closest to the inspection region in the direction in which the identification document is inserted. This prevents the accuracy of authenticity determination from being reduced due to the influence of ambient light when photographing identification documents.
[0014] The inspection area and the base material area are set for each type of identification document, and it is also desirable that the acquisition means cut out the inspection area and the base material area according to the type of identification document. By defining the appropriate cutout positions for inspection areas and base material areas for each type of identification document, and cutting out the inspection areas and base material areas according to the type of identification document to be judged, it is possible to provide an authenticity determination device that can handle multiple types of identification documents.
[0015] The second invention is a program for causing a computer to function as a counterfeit determination device for determining the authenticity of an identification document, comprising: an acquisition means for obtaining a UV inspection image and a UV substrate image, respectively, by cutting out the inspection area and the substrate area of the identification document from a UV image taken under UV (Ultra Violet) illumination of the identification document to be determined; and an authenticity determination means for determining the range of the substrate color of the identification document in the UV image using the UV substrate image, calculating the area of the non-substrate portion that has different fluorescence characteristics from the substrate from the UV inspection image using that range, and determining the authenticity of the identification document to be determined by comparing that area with a threshold. The second invention is a program for the authenticity determination device of the first invention. [Effects of the Invention]
[0016] The present invention makes it possible to provide a genuineness determination device, etc., that can suitably determine the authenticity of an identity verification document. [Brief explanation of the drawing]
[0017] [Figure 1] A diagram illustrating the authenticity determination system 1. [Figure 2] A diagram showing the camera 21 and lighting 22 inside scanner 2. [Figure 3] A diagram showing the hardware configuration of the authenticity determination device 3. [Figure 4] Figure showing the ID card 10. [Figure 5] Figure showing the functions of the genuine / fake determination device 3. [Figure 6] Flowchart showing the outline of the genuine / fake determination method. [Figure 7] Figure for explaining the photographing of the ID card 10. [Figure 8] Figure for explaining the orientation when the ID card 10 is inserted into the scanner 2. [Figure 9] Flowchart showing the procedure for genuine / fake determination based on the fluorescence characteristics of the base material 11. [Figure 10] Figure showing the inspection regions Rv, Ru and the base material region Rb of the ID card 10, and the visible light inspection image 110, the UV inspection image 120, and the UV base material image 130. [Figure 11] Example of the histogram of the UV base material image 130. [Figure 12] Flowchart showing the procedure for extracting the base material part. [Figure 13] Figure for explaining the extraction of the base material part. [Figure 14] Figure for explaining noise removal. [Figure 15] Flowchart showing the procedure for extracting the character part. [Figure 16] [[ID= 37]]Figure for explaining the extraction of the character part. [Figure 17] Figure for explaining the creation of the composite image 400.
Embodiments for Carrying Out the Invention
[0018] Hereinafter, preferred embodiments of the present invention will be described in detail based on the drawings.
[0019] (1. Genuine / Fake Determination System 1) FIG. 1 is a diagram showing a genuine / fake determination system 1 having a genuine / fake determination device 3 according to an embodiment of the present invention. As shown in FIG. 1, the genuine / fake determination system 1 is configured by connecting a scanner 2 and a genuine / fake determination device 3 communicably by wire or wirelessly.
[0020] In the authenticity verification system 1, a photograph of the ID card 10 is transmitted from the scanner 2 to the authenticity verification device 3, which then performs authenticity verification of the ID card 10 based on the photograph. The purpose of the authenticity verification is not particularly limited and can be applied to various purposes, including opening an account at a financial institution, signing a contract with a mobile carrier, and verifying identity during procedures at various government agencies.
[0021] Scanner 2 is used to photograph the ID card 10. As shown in Figure 2, a camera 21 and a light 22 are provided inside scanner 2. Scanner 2 photographs the ID card 10, which is inserted through the insertion slot A (see Figure 1) and set in a predetermined position, with the camera 21 located directly above it. The light 22 is located diagonally above both sides of the ID card 10.
[0022] Camera 21 is an area camera composed of an optical lens, an image sensor such as a CCD (Charge Coupled Device) or CMOS (Complementary Metal-Oxide Semiconductor), and an A / D (Analog / Digital) conversion unit.
[0023] In this embodiment, the illumination 22 is provided with visible light illumination 22a and UV (Ultra Violet) illumination 22b. For example, an LED (Light Emitting Diode) is used as the visible light illumination 22a, but is not limited to this.
[0024] In this embodiment, an IC chip is provided inside the ID card 10, and type identification information indicating the type of ID card 10, such as a driver's license, My Number card, residence card, or special permanent resident certificate, is stored in this IC chip. In addition to the camera 21 and illumination 22, the scanner 2 of this embodiment is equipped with a card reader (reading device) (not shown) for reading the type identification information in the IC chip.
[0025] Figure 3 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 storage unit 32, a communication unit 33, a display unit 34, etc., via a bus or the like. However, it is not limited to this, and various configurations can be adopted as appropriate.
[0026] The control unit 31 consists of a CPU, ROM, RAM, etc. The CPU calls and executes programs related to the processing of the authenticity determination device 3, which are stored in storage media such as the memory unit 32 and ROM, into a work area on the RAM. ROM is a non-volatile memory that permanently holds programs and data such as the boot program and BIOS. RAM is a volatile memory that temporarily holds 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.
[0027] The memory unit 32 is a hard disk drive, solid-state drive, flash memory, etc., and stores the program that the authenticity determination device 3 executes during the processing described later, the data necessary for program execution, the OS, etc.
[0028] The communication unit 33 is a communication interface that mediates communication with the scanner 2. The display unit 34 is a liquid crystal display or the like, and displays various information related to the authenticity determination process of this embodiment.
[0029] (2. ID card 10) In this embodiment, ID card 10 is an identification document that is subject to authenticity verification. Identification documents are media used for administrative procedures and identity verification by various government agencies, financial institutions, mobile carriers, etc., and in this embodiment, it is specifically a driver's license. However, ID card 10 is not limited to this and may also be a My Number Card, residence card, special permanent resident certificate, etc.
[0030] Figure 4 is a schematic diagram of ID card 10 (driver's license), showing a simplified representation with some details omitted from the actual card surface. Various information is printed on the rectangular base material 11 of ID card 10. For example, for a driver's license, textual information such as the person's name, date of birth and address, the date of issuance, number, and expiration date are printed in designated locations. Image information such as the person's facial image and various lines are also printed in designated locations. The composition of the information on ID card 10 varies depending on the type of ID card, such as a driver's license, My Number card, residence card, or special permanent resident certificate.
[0031] (3. Authenticity Determination Device 3 Functions) Figure 5 is a block diagram showing the functions of the authenticity determination device 3. The authenticity determination device 3 includes an acquisition means 301, an authenticity determination means 302, etc.
[0032] The acquisition means 301 is used by the control unit 31 of the authenticity determination device 3 to acquire images used for authenticity determination (visible light inspection image 110, UV inspection image 120, and UV substrate image 130, which will be described later) from images taken of the ID card 10 under visible light illumination 22a and UV illumination 22b. These images will be described later.
[0033] The authenticity determination means 302 is a mechanism in which the control unit 31 of the authenticity determination device 3 uses the above image to determine the authenticity of the ID card 10. This authenticity determination will be described later.
[0034] (4. Methods for determining authenticity) Figure 6 is a flowchart illustrating the authenticity determination method performed by the authenticity determination system 1. S1 to S3 in Figure 6 are processes executed by the control unit (not shown) of the scanner 2, which controls each part of the scanner 2, and S4 to S9 are processes executed by the control unit 31 of the authenticity determination device 3, which controls each part of the authenticity determination device 3.
[0035] In this embodiment, first the user inserts the ID card 10 into the scanner 2 and sets it in a predetermined position, and the scanner 2 takes a picture of the ID card 10 with the camera 21 in response to the user's operation (S1).
[0036] In this embodiment, when photographing the ID card 10, the scanner 2 first photographs the ID card 10 with only the visible light illumination 22a lit, as shown in Figure 7, and then photographs the ID card 10 with only the UV illumination 22b lit. In this way, the visible light illumination 22a and UV illumination 22b are individually lit and photographed to obtain a pair of captured images. In each captured image, the color of the pixels is represented by the pixel values of three channels: R (red), G (green), and B (blue). Here, "color" includes not only hue but also the element of light and dark.
[0037] Scanner 2 further reads the ID card type identification information stored in the IC chip of ID card 10 (S2). Then, it transmits the pair of captured images obtained by photographing ID card 10 in S1 and the ID card type identification information of ID card 10 read in S2 (referred to as captured images of ID card 10, etc.) to the authenticity determination device 3 (S3).
[0038] When the authenticity determination device 3 receives the captured image of the ID card 10 (S4), it uses the captured image of the ID card 10 taken under visible light illumination 22a (hereinafter referred to as the visible light captured image) from the pair of captured images to detect the orientation of the ID card 10 (S5).
[0039] In other words, in this embodiment, as shown in Figure 8(a), the ID card 10 is inserted into the scanner 2 with its upper side facing inward. At this time, by photographing the ID card 10 under visible light illumination 22a, a visible light image is obtained with the upper side of the ID card 10 facing upward, as shown in the left diagram of Figure 8(b).
[0040] On the other hand, if the ID card 10 is inserted incorrectly, a visible light image will be obtained with the bottom of the ID card 10 facing upwards, as shown in the right-hand diagram of Figure 8(b). Therefore, if the ID card 10 is inserted incorrectly, it can be detected as an error by performing OCR (optical character recognition) on the visible light image of the ID card 10 to detect the orientation of the characters.
[0041] If the ID card 10 is oriented incorrectly (S6; NO), the authenticity determination device 3 displays an error message on the display unit 34 (S7) and terminates the process. The person in charge can then take appropriate action based on this display.
[0042] If the ID card 10 is oriented correctly (S6; YES), the authenticity determination device 3 uses the pair of captured images received in S4 to determine the authenticity of the ID card 10 (S8). Details of the authenticity determination will be described later.
[0043] The authenticity determination device 3 displays the result of the authenticity determination of the ID card 10 on the display unit 34 (S9), and then terminates the process. The person in charge can take appropriate action according to the displayed content.
[0044] (5. Authenticity determination based on the fluorescence properties of the substrate 11) In this embodiment, the authenticity of the ID card 10 is determined by utilizing the fact that the base material 11 of the ID card 10 has predetermined fluorescence characteristics depending on the amount of fluorescent agent it contains. That is, if an alteration is performed by attaching a piece of paper or the like with different fluorescence characteristics from the base material 11 to the ID card 10, the amount of fluorescence (fluorescence intensity) under UV illumination 22b will change between the base material 11 of the ID card 10 and the attached material, so such an ID card 10 can be suitably detected as a counterfeit ID card 10.
[0045] The procedure for determining authenticity based on the fluorescence properties of the substrate 11 will be described below with reference to Figure 9, etc. Figure 9 is a flowchart showing the procedure for determining authenticity in S8 of Figure 6.
[0046] In this embodiment, in S8, first, the area of the ID card 10 is extracted from the captured image (hereinafter referred to as the UV image) and the visible light image of the ID card 10 taken under UV illumination 22b, which were received in S4 (see Figure 6), and a predetermined area of the ID card 10 is cut out from that area (S801).
[0047] In this embodiment, the predetermined regions to be cut out in S801 are set in advance for both the UV image and the visible light image. Figure 10(a) shows examples of the predetermined regions Rv, Ru, and Rb in the visible light image and the UV image.
[0048] The predetermined area Rv is defined in the visible light image as the area to be inspected where alteration due to the attachment of an object is expected. It is mainly set in the part of the ID card 10 that contains different text information (e.g., name, address, ID card number, etc., also called on-demand information), even if the ID cards 10 are of the same type, but it can also be set in the part of other text information. The predetermined area Ru is defined in the UV image as the area at the same position as the predetermined area Rv, and these predetermined areas Rv and Ru will be referred to as the inspection area below. The predetermined area Rb is defined in the UV image as the area of the base material 11 of the ID card 10, and is set to avoid the printed area of the card surface information. Hereinafter, this predetermined area Rb will be referred to as the base material area.
[0049] Multiple inspection areas Rv and Ru are defined on the ID card 10. The numbers in parentheses following the codes Rv and Ru in Figure 10(a) are sequential numbers assigned to multiple inspection areas Rv and Ru, and the same sequential number indicates the same location. For example, inspection areas Rv(1) and Ru(1) are set to the same location in both the visible light and UV images. Similarly, multiple substrate areas Rb are defined, and the Roman numerals in parentheses following the code Rb in Figure 10(a) are sequential numbers assigned to multiple substrate areas Rb.
[0050] The positions of the inspection areas Rv and Ru and the substrate area Rb are predetermined for each type of ID card 10 and stored in the memory unit 32 of the authenticity determination device 3. In S801, the type identification information received in S4 (see Figure 6) is referenced to extract the inspection areas Rv and Ru and the substrate area Rb according to the type of ID card 10. Hereinafter, the image obtained by extracting the inspection area Rv from the visible light image will be called the visible light inspection image, and the images obtained by extracting the inspection area Ru and the substrate area Rb from the UV image will be called the UV inspection image and the UV substrate image, respectively.
[0051] Figure 10(b) shows examples of visible light inspection image 110, UV inspection image 120, and UV substrate image 130 obtained by cutting out the inspection area Rv(1), Ru(1), and substrate area Rb(i) from Figure 10(a). In this example, an attachment S is attached to the inspection area Ru(1) (=Rv(1)) of the ID card 10, and the fluorescence properties are different between the attachment S and the substrate 11.
[0052] The authenticity determination device 3 resizes the visible light inspection image 110, the UV inspection image 120, and the UV substrate image 130 to a predetermined size (S802). This speeds up the processing from S803 onward.
[0053] Next, the authenticity determination device 3 determines the color range of the substrate 11 in the UV image from the UV substrate images 130 of each substrate region Rb(i) to Rb(iii) in the UV image (S803).
[0054] In S803, for each substrate region Rb(i) to Rb(iii), a histogram showing the distribution of the number of pixels for each pixel value is created for each RGB channel of the UV substrate image 130. Figure 11 is an example of the histogram for the G channel, where the horizontal axis is the pixel value and the vertical axis is the number of pixels for each pixel value.
[0055] Next, the number of peaks P in the histogram is determined, and the value obtained by multiplying the number of peaks P by a predetermined coefficient th_rangeMatl is set as the reference value P × th_rangeMatl. Then, the pixel values V1 and V2 located immediately on either side of the pixel value Vp corresponding to the number of peaks P, which are the pixel values whose number of pixels is the reference value P × th_rangeMatl, are determined. The coefficient th_rangeMatl is predetermined to an appropriate value and stored in the memory unit 32 of the authenticity determination device 3.
[0056] In this embodiment, the range between these pixel values V1 and V2 is defined as the color range of the substrate 11 in the G channel. The color range of the substrate 11 is also determined for the R channel and B channel using the same method as described above, and in S803, the color range of the substrate 11 for each RGB channel is determined for each of the UV substrate images 130 from substrate region Rb(i) to Rb(iii). Note that the method for calculating the color range of the substrate 11 is not limited to this, and any method that can calculate an appropriate range as the color of the substrate 11 is acceptable. For example, it is also possible to define a certain range centered on the aforementioned pixel value Vp as the color range of the substrate 11.
[0057] Next, the authenticity determination device 3 uses the color range of the substrate 11 obtained in S803 to extract the substrate portion from the UV inspection images 120 of each inspection area Ru(1) to Ru(5) of the UV captured image (S804).
[0058] Figure 12 is a flowchart showing a specific example of the processing in S804. In this embodiment, in S804, first, R, G, and B images are created that show the pixel values of each RGB channel of the UV inspection image 120 (S8041). Figure 13(a) is an example of the R, G, and B images obtained from the UV inspection image 120 in Figure 10(b), and is a grayscale image that shows the high and low pixel values of each RGB channel in grayscale.
[0059] Subsequently, these R, G, and B images are binarized into a substrate portion where the pixel values fall within the color range of the substrate 11 determined in S803, and a non-substrate portion where the pixel values fall outside that range (S8042).
[0060] This process is performed for each of the UV inspection images 120 for each inspection area Ru(1) to Ru(5). In this process, the color range of the substrate 11 should be calculated based on the substrate area Rb that is closest to the target inspection area Ru in the direction in which the ID card 10 is inserted into the scanner 2.
[0061] In other words, when binarizing the R, G, and B images of the UV inspection image 120 extracted from the inspection area Ru(1) in Figure 10(a), the color range of the substrate 11 is determined using the color range of the R, G, and B channels obtained from the UV substrate image 130 of the substrate area Rb(i) that is closest to the inspection area Ru(1) in the insertion direction of the ID card 10.
[0062] This is because, if the light-shielding cover (not shown) of the scanner 2 is not lowered when the ID card 10 is being photographed, there is a risk that ambient light entering from the insertion slot A of the scanner 2 may affect the UV image. However, even in that case, if the insertion direction of the ID card 10 is approximately the same for the inspection area Ru and the substrate area Rb in the UV image, the influence of ambient light on the inspection area Ru and the substrate area Rb will be approximately the same, and the color of the substrate 11 in the UV image can be expected to be approximately the same between the inspection area Ru and the substrate area Rb.
[0063] Figure 13(b) shows an example of extracting the substrate portion by binarizing the R, G, and B images, where the substrate portion whose pixel value falls within the color range of the substrate 11 is shown as white pixels, and the non-substrate portion whose pixel value falls outside that range is shown as black pixels. In this embodiment, the binary images of the R, G, and B images are inverted in black and white, and the resulting binary images are combined (S8043). Figure 13(c) shows an example of the binary images after black and white inversion, and Figure 13(d) shows the combined image 210 obtained by combining these images.
[0064] The method for combining binary images is not particularly limited, but in this embodiment, pixels at the same position are compared for the (black and white inverted) binary images of the R image, G image, and B image. If all pixels are black, the pixels at that position in the combined image 210 are made black; otherwise, the pixels at that position in the combined image 210 are made white. By performing this process for pixels at all positions, the combined image 210 shown in Figure 13(d) is obtained.
[0065] In the composite image 210, the substrate portion appears as black pixels. However, in this embodiment, as shown in Figure 13(e), known morphological transformation and noise reduction processing are performed (S8044) to reshape the substrate portion and obtain the substrate portion extracted image 230.
[0066] In this embodiment, the morphological transformation described above involves performing opening, closing, and shrinking processes in that order for the region of white pixels. The opening process shrinks the region of white pixels and then expands it, while the closing process expands the region of white pixels and then shrinks it. The expansion and shrinking processes are known and will not be explained further.
[0067] Noise reduction processing is performed to avoid the influence of characters or background patterns when they are present on the base material 11 of the ID card 10. In this respect, the noise reduction method is not particularly limited, but in this embodiment, as shown in Figure 14(a), regions W (W1 to W4) in which a certain number (for example, 8 pixels) of white pixels are consecutive are extracted from the composite image 210 after morphological transformation. Then, a number of continuous region division images 220 are created, each the same size as the composite image 210, with all pixels other than each region W1 to W4 being black, for the number of regions W1 to W4.
[0068] Subsequently, only continuous region segmentation images 220 having a region W of a predetermined area or larger are retained. If there are multiple remaining continuous region segmentation images 220, these images are combined to create a substrate portion extraction image 230. In the example shown in Figure 14(a), the remaining continuous region segmentation images 220 are only those containing region W1, and these images 220 become the substrate portion extraction image 230.
[0069] The above processing is performed on each of the UV inspection images 120 in inspection areas Ru(1) to Ru(5). The predetermined area is determined for each inspection area Ru(1) to Ru(5) based on the rectangular area (font size) per character (including numbers, symbols, etc.) within the inspection areas Ru(1) to Ru(5) of the regular ID card 10.
[0070] The predetermined area is, for example, the value obtained by reducing the rectangular area by the amount of resizing in S802, and then multiplying the number of pixels for the reduced area by a predetermined coefficient. This predetermined area is linked to the positions of the inspection areas Ru(1) to Ru(5) and is stored in advance in the memory unit 32 of the authenticity determination device 3.
[0071] Furthermore, when synthesizing the remaining continuous region segmented images 220, for example, pixels at the same position are compared among the remaining continuous region segmented images 220, and as described above, if all pixels are black, the pixels at that position in the substrate portion extraction image 230 are set to black; otherwise, the pixels at that position in the substrate portion extraction image 230 are set to white. Note that if there is no region W with a certain number of consecutive white pixels in the synthesized image 210 after morphological transformation, or if there is no continuous region segmented image 220 with a region W of a predetermined area or larger, an image of the same size as the synthesized image 210, with all pixels set to black, is created as the substrate portion extraction image 230. However, the synthesis method for the continuous region segmented images 220 is not limited to this.
[0072] As described above, the substrate portion is extracted from each of the UV inspection images 120 of the inspection areas Ru(1) to Ru(5) of the UV image. In the substrate portion extraction image 230, the substrate portion is shown as black pixels, and if an attachment with different fluorescence properties from the substrate 11 of the ID card 10 is attached to the inspection areas Ru(1) to Ru(5), the attachment is shown as a non-substrate portion with white pixels.
[0073] Figure 14(b) shows an example of applying the above noise reduction processing to a composite image 210 after morphological transformation of a genuine ID card 10 that has not been altered by the attachment of the attached object S. In the composite image 210, the character portions on the substrate 11 appear as non-substrate portions (white pixels), but by performing the above noise reduction processing, these character portions no longer appear as non-substrate portions (white pixels) in the substrate portion extraction image 230, and the character portions on the substrate 11 are no longer treated as non-substrate portions. In S807, which will be described later, authenticity is determined by comparing the area of the non-substrate portions with a threshold, but by performing the above noise reduction processing, it is possible to prevent the character portions on the substrate 11 from becoming non-substrate portions, thereby preventing the genuine ID card 10 from being mistakenly detected as a fake ID card 10.
[0074] Returning to the explanation of the flowchart in Figure 9, the authenticity determination device 3 then extracts the text portion from the visible light inspection images 110 of each inspection area Rv(1) to Rv(5) of the visible light captured image (S805).
[0075] Figure 15 is a flowchart illustrating a specific example of the processing in S805. In this embodiment, in S805, first, R, G, and B images are created that show the pixel values of each RGB channel of the visible light inspection image 110 (S8051). Figure 16(a) is an example of the R, G, and B images obtained from the visible light inspection image 110 in Figure 10(b), and is a grayscale image that shows the high and low pixel values of each RGB channel in grayscale.
[0076] Subsequently, these R, G, and B images are binarized (S8052). For binarization, Otsu's method, which automatically determines the binarization threshold according to the image, can be used. Otsu's method is well-known and will not be explained. Figure 16(b) shows an example of binarized R, G, and B images, where smaller pixel values are shown as black pixels, larger pixel values as white pixels, and text is extracted as black pixels. Note that the binarization method is not limited to the above; it may also be performed based on a predetermined threshold.
[0077] In this embodiment, the binary images of the R image, G image, and B image are combined (S8053) to create a composite image 310 as shown in Figure 16(c). The method for combining binary images is not particularly limited, but in this embodiment, pixels at the same position in the binary images of the R image, G image, and B image are compared, and if any pixel is black, the pixel at that position in the composite image 310 is made black, and if all pixels are white, the pixel at that position in the composite image 310 is made white. By performing this process for pixels at all positions, the composite image 310 shown in Figure 16(c) is obtained.
[0078] In the composite image 310, the character portion is represented by black pixels. However, in this embodiment, as shown in Figure 16(d), a known morphological transformation is performed (S8054) to reshape the character portion and obtain the character portion extracted image 320. In this embodiment, as the morphological transformation, a known shrinkage process is performed on the region of white pixels.
[0079] Returning to the explanation of the flowchart in Figure 9, the authenticity determination device 3 uses the substrate portion extraction image 230 created in S804 and the character portion extraction image 320 created in S805 to determine the area of the non-substrate portion (S806).
[0080] In this embodiment, in S806, the substrate portion extraction image 230 and the character portion extraction image 320 obtained from inspection areas Ru and Rv at the same location are combined. For example, in the example of Figure 17(a), the substrate portion extraction image 230 obtained from inspection area Ru(1) of the UV image in Figure 10(a) and the character portion extraction image 320 obtained from inspection area Rv(1) of the visible light image are combined to obtain a combined image 400. The authenticity determination device 3 calculates the area of the white pixel region in the combined image 400 as the area of the non-substrate portion, based on the number of pixels.
[0081] The synthesis method is not particularly limited, but in this embodiment, the pixels at the same position in the substrate portion extracted image 230 and the character portion extracted image 320 are compared, and as described above, if any of the pixels are black, the pixels at that position in the synthesized image 400 are made black, and if all of the pixels are white, the pixels at that position in the synthesized image 400 are made white.
[0082] As a result, the area of the character portion is excluded from the area of the non-substrate portion, and the character portion on the non-substrate portion is also excluded from the area of the non-substrate portion. Furthermore, as shown in Figure 17(b), even if the character portion on the substrate 11 of the ID card 10 remains as a non-substrate portion (white pixels) in the substrate portion extraction image 230 for some reason, such as a color pattern C superimposed on the character portion on the substrate 11, the area of the non-substrate portion can be minimized by the synthesis process, thereby reducing its impact on authenticity determination. Decomposing the visible light inspection image 110 into R image, G image, and B image in S805 is also effective in that the above-mentioned color pattern C is revealed in the character portion extraction image 320 and its influence is excluded in the synthesized image 400.
[0083] The above process is performed for all of the multiple inspection areas Ru(1) to Ru(5) (=Rv(1) to Rv(5)) on the ID card 10, and the area of the non-substrate portion is calculated for each of these inspection areas Ru(1) to Ru(5). The authenticity determination device 3 compares each area with a threshold, and if any area exceeds the threshold (S807; NO), it determines that the ID card 10 is fake, assuming that an alteration has occurred in the corresponding inspection area Ru by attaching an object S to the ID card 10 (S809), and terminates the process.
[0084] On the other hand, if all areas are below the threshold (S807; YES), the authenticity determination device 3 determines that the ID card 10 to be determined is true (S808) and terminates the process. The above thresholds are set for each inspection area Ru(1) to Ru(5) and are stored in advance in the storage unit 32 of the authenticity determination device 3, linked to the position of each inspection area Ru(1) to Ru(5).
[0085] As described above, in this embodiment, the authenticity of the ID card 10 is determined from a UV image taken of the ID card 10 under UV illumination 22b, utilizing the fact that the base material 11 of the ID card 10 has predetermined fluorescence properties. This makes it possible to detect counterfeit ID cards 10 that have been altered by printing false information on a piece of paper or the like with different fluorescence properties from the base material 11 of the ID card 10 and attaching it. This determination method is resistant to scratches and dirt on the base material 11, and the accuracy of authenticity determination does not decrease significantly even if scratches or dirt occur on the surface of the ID card 10.
[0086] Furthermore, in this embodiment, the noise reduction process is performed when creating the substrate portion extraction image 230. This prevents the characters on the substrate 11 of the ID card 10 from being extracted as non-substrate portions, thereby preventing a genuine ID card 10 from being mistakenly detected as a fake ID card 10.
[0087] Furthermore, in this embodiment, as described above, by using the character portion extraction image 320, the area of the non-substrate portion can be calculated by excluding the character portion, and the area of the non-substrate portion can be calculated by excluding the character on the non-substrate portion. In addition, even if the character on the substrate 11 remains as the non-substrate portion of the substrate portion extraction image 230 for some reason, the influence of the character on the authenticity determination can be reduced.
[0088] Furthermore, in this embodiment, when creating the substrate portion extraction image 230, the color range of the substrate 11 obtained from the UV substrate image 130 of the substrate region Rb closest to the inspection region Ru of the UV inspection image 120 in the insertion direction of the ID card 10 is used. This prevents a decrease in the accuracy of authenticity determination due to the influence of ambient light when the ID card 10 is photographed.
[0089] Furthermore, in this embodiment, the cutting positions of the inspection areas Rv and Ru and the base material area Rb, which are suitable for authenticity determination, are determined for each type of ID card 10, and by cutting out the area corresponding to the type of ID card 10 to be determined, an authenticity determination device 3 that can handle multiple types of ID cards 10 can be provided.
[0090] However, the present invention is not limited to the above embodiments. For example, in this embodiment, the type of ID card 10 to be judged can be identified by type identification information within the IC chip, but it is also possible to identify the type of ID card 10 to be judged from a visible light image of the ID card 10. In this case, instead of receiving type identification information of the ID card 10 from the scanner 2, the authenticity determination device 3 can determine the type of ID card 10 from a visible light image of the ID card 10.
[0091] In this embodiment, as shown in Figure 10(a), the insertion direction of the ID card 10 into the scanner 2 is along the short side of the ID card 10, but this is not limited to this, and the insertion direction of the ID card 10 may be along the long side of the ID card 10. In this case as well, when extracting the substrate portion of the UV inspection image 120, the color range of the substrate 11 obtained from the UV substrate image 130 of the substrate region Rb closest to the inspection region Ru of the UV inspection image 120 in the insertion direction of the ID card 10 (the direction of the long side of the ID card 10) is used.
[0092] Furthermore, in this embodiment, when extracting the substrate portion of the UV inspection image 120, the color range of the substrate 11 is calculated using the substrate region Rb closest to the inspection region Ru of the UV inspection image 120 in the insertion direction of the ID card 10, but it is not limited to this. In some cases, it is also possible to use the calculation using the substrate region Rb second closest to the inspection region Ru.
[0093] Furthermore, the method for calculating the area of the non-substrate portion is not limited to the above; any method that uses the color range of the substrate 11 obtained from the UV substrate image 130 and defines pixels of colors outside that range as the non-substrate portion in the UV inspection image 120 is acceptable. For example, in some cases, the noise reduction process in S8044 (see Figure 12) can be omitted, and the character portion extraction process in S805 (see Figure 9) can be omitted, and the area of the non-substrate portion in the substrate portion extraction image 230 can be used directly for determination.
[0094] Preferred embodiments of the present invention have been described above with reference to the attached drawings, but the present invention is not limited to these examples. It will be obvious to those skilled in the art that various modifications or alterations can be conceived within the scope of the technical idea disclosed herein, and these will naturally also fall within the technical scope of the present invention. [Explanation of Symbols]
[0095] 1: Authenticity Verification System 2: Scanner 3: Authenticity verification device 10: ID card 21: Camera 22: Lighting 22a:Visible light illumination 22b:UV lighting 11: Base material 110: Visible light inspection image 120: UV inspection image 130: UV substrate image 301: Acquisition method 302: Means for determining authenticity
Claims
1. A device for determining the authenticity of an identification document, An acquisition method for obtaining UV inspection images and UV substrate images, respectively, by extracting the inspection area and substrate area of the identification document from a UV image taken under UV (Ultra Violet) illumination of the identification document to be judged, and A genuineness determination means that determines the authenticity of the identity document to be determined by using the aforementioned UV substrate image to determine the range of the substrate color of the identity document in the UV photographed image, using that range to calculate the area of the non-substrate portion that has different fluorescence properties from the substrate from the UV inspection image, and comparing that area with a threshold, A device for determining authenticity, characterized by having [a certain feature].
2. The authenticity determination means creates a substrate portion extraction image from the UV inspection image showing the substrate portion and the non-substrate portion in order to calculate the area. The authenticity determination device according to claim 1, characterized in that when creating the extracted image of the substrate portion, noise reduction processing is performed so that the character portion on the substrate is not treated as the non-substrate portion.
3. The acquisition means acquires a visible light inspection image by cutting out the inspection area from a visible light image taken of the identity verification document under visible light illumination, The authenticity determination device according to claim 1, characterized in that the authenticity determination means creates a character portion extraction image by extracting the character portion from the visible light inspection image in order to calculate the area, and calculates the area excluding the character portion as the area.
4. The aforementioned UV image was taken under UV illumination of an identification document inserted into a scanner. Multiple substrate regions are set, The authenticity determination device according to claim 1, characterized in that the range of the substrate color is determined from the UV substrate image of the substrate region closest to the inspection area in the direction in which the identification document is inserted.
5. The inspection area and the substrate area are set according to the type of identification document. The authenticity determination device according to claim 1, characterized in that the acquisition means cuts out the inspection area and the base material area according to the type of identity verification document.
6. Computers, A device for determining the authenticity of an identification document, An acquisition method for obtaining UV inspection images and UV substrate images, respectively, by extracting the inspection area and substrate area of the identification document from a UV image taken under UV (Ultra Violet) illumination of the identification document to be judged, and A genuineness determination means that determines the authenticity of the identity document to be determined by using the aforementioned UV substrate image to determine the range of the substrate color of the identity document in the UV photographed image, using that range to calculate the area of the non-substrate portion that has different fluorescence properties from the substrate from the UV inspection image, and comparing that area with a threshold, A program for functioning as a genuine / counterfeit detection device, characterized by having [a certain feature].