JP2007279812A - Authenticity determination system, authenticity determination method, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an authenticity determination system in which both convenience and security are enhanced. <P>SOLUTION: A determination server 2 has: a reference data storage section 202 for storing, with components therefor, a reference pattern composed of random patterns (e.g. fiber pattern of paper) of an object surface; a correlation value computing section 204 for computing a correlation value between the stored reference pattern and a determination object pattern (i.e. image data transmitted with a PFP acquisition device 10); a threshold control section 206 for controlling a control value; and a comparison section 208 for comparing the computed correlation value with a controlled threshold to determine identity of the reference pattern and the determination object pattern. The threshold control section 206 controls the threshold, based on a determination history and an amount and the number of commodities at the use of a card. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a true / false determination system for performing true / false determination of a credit card, a cash card or the like.

  In this type of authenticity judgment system, the error rate (FRR: False Accept Rate) that accepts a genuine article as a fake and the error rate (FAR; False Accept Rate) that accepts a genuine article as a genuine article are convenient and safe. Are related to sex and are in a trade-off relationship with each other. If the determination threshold is made stricter in order to eliminate counterfeits with an emphasis on safety, the rate at which genuine items are excluded increases and convenience deteriorates. On the other hand, if importance is placed on convenience, safety is impaired.

Patent Documents 1 to 6 are examples of techniques for performing authenticity determination.
Patent Document 1 discloses a method for determining the authenticity of a card-like medium itself and the authenticity of a user. However, the determination result is only a true or false binary value, and no pattern matching is disclosed.
Patent Document 2 discloses a method in which a plurality of pieces of information for a registrant are written on an IC card, and identity verification is determined using a plurality of pieces of biometric information according to the relevance rate when the card is used. However, this approach reduces FAR but does not improve FRR.

  In Patent Document 3, a fingerprint of a registrant is registered in an ID card, and at the time of fingerprint collation, a fingerprint input unit, input fingerprint data creation unit, or collation determination unit is adjusted according to fingerprint processing information recorded on the ID card Is disclosed. However, this method only has different verification determination means for each card.

  In Patent Document 4, personal authentication information relating to a user's facial feature amount and judgment level information relating to a similarity distribution between the facial features of the user's face and another person's face are recorded, and FRR / FAR information obtained by a judgment threshold value Discloses a face image authentication technique for determining permission of a predetermined process according to a security level. However, in this method, it is necessary to determine in advance a judgment level regarding the distribution of similarity between the face of the person and the other person.

  Patent Document 5 discloses a face authentication technique for determining whether or not to update face image data recorded on a card based on a collation result, and corresponding to the temporal change of the person's face. However, with this method, if the collation result decreases, it is unclear whether the cause is similar to another person's face or the person's face over time, so register with another person's face. There is a possibility that.

  In Patent Document 6, two threshold values for fingerprint authentication are prepared. If the value is higher than the higher one, the use is permitted. If the value is lower than the lower one, the credit card authentication technique is performed. It is disclosed. However, this method performs a binary determination that permits or disallows the use regardless of the usage situation of the card.

  Patent Document 7 discloses a true / false determination method using a so-called paper fingerprint technique.

JP 2000-163553 A JP 2001-338295 A JP 2003-67745 A JP 2003-99780 A JP 2003-337926 A JP-A-4-157586 JP 2005-38389 A

  The present invention has been made from the above-described background, and an object thereof is to provide a true / false determination system capable of enhancing both convenience and safety.

  In order to achieve the above object, a true / false determination system according to the present invention includes a storage unit that stores a reference pattern composed of a random pattern on an object surface, and a reference pattern stored in the storage unit and a determination target pattern. Based on the calculation means for calculating the correlation value, the threshold value control means for controlling the threshold value, the correlation value calculated by the calculation means and the threshold value controlled by the threshold value control means, the reference pattern and the determination target pattern are the same Determination means for determining sex.

Preferably, the threshold control means controls the threshold based on a determination history stored in advance.
Preferably, the apparatus further includes an accepting unit that accepts an amount, and the threshold control unit controls the threshold based on the amount accepted by the amount accepting unit.
Preferably, the reception unit further receives the number of products, and the threshold control unit controls the threshold based on the number of products received by the reception unit.

Further, the authenticity determination method according to the present invention stores a reference pattern composed of a random pattern on the object surface, calculates a correlation value between the stored reference pattern and a determination target pattern,
A threshold value is controlled, and based on the calculated correlation value and the controlled threshold value, the identity between the reference pattern and the determination target pattern is determined.

  The program according to the present invention, in a true / false determination system including a computer, stores a reference step consisting of a random pattern on the surface of an object, and a correlation value between the stored reference pattern and the determination target pattern. A calculation step for calculating, a threshold control step for controlling a threshold, and a determination step for determining the identity of the reference pattern and the determination target pattern based on the calculated correlation value and the controlled threshold. Have the fake decision system computer run it.

  According to the authenticity determination system of the present invention, both convenience and safety can be improved.

First, a pattern matching card used in the authenticity determination system of the present invention will be described.
1A and 1B are diagrams illustrating a pattern matching card. FIG. 1A illustrates a fiber pattern of paper, FIG. 1B illustrates a pattern matching card made of paper, and FIG. ) Indicates a pattern matching card made of plastic or silver paint.
As illustrated in FIG. 1A, the paper is made of intricately intertwined plant fibers. The width of the plant fiber is generally 4 to 70 μm, 10 to 50 μm for plain paper made from hardwood, and 20 to 70 μm for plain paper made from conifer.

  The use of a random pattern such as plant fibers contained in paper as a “fingerprint” for identifying each paper is also referred to as a PFP (Paper Finger Print) technique. Note that the PFP technique can be applied not only to paper but also to an object having a random pattern, such as a blasted metal plate or a silver-coated plastic plate.

  The pattern matching card includes a random pattern thus naturally formed as an identification target. As shown in FIGS. 1B and 1C, the pattern matching card has a random pattern of paper, plastic, or silver paint. The pattern matching card shown in FIG. 1B has a paper fiber pattern as a random pattern. The random pattern portion to be identified is included in at least a part of the pattern matching card, and its area is 1 to 2 mm square. Since it is sufficient that a 1 to 2 mm square plane area of the identification target portion is secured in any part of the card key, the degree of freedom in designing the entire card becomes very high, and the card can be easily downsized.

  The pattern matching card is determined to be genuine or fake by a true / false determination system described later. Since it is difficult to intentionally reproduce the paper fiber pattern, each of the above-mentioned pattern matching cards can be identified with extremely high accuracy, and its forgery is extremely difficult.

Next, the authenticity determination system 1 according to the embodiment of the present invention will be described.
FIG. 2 is a diagram showing the authenticity determination system 1 that performs the authenticity determination process of the pattern matching card 6.
As shown in FIG. 2, the authenticity determination system 1 includes a PFP acquisition device 10 that acquires image data of the PFP registration unit 8 from a pattern matching card 6 having a PFP registration unit 8, and the PFP acquisition device 10 and the network 100. And a determination server 2 connected via the network. The determination server 2 is connected to, for example, a history server 3 of a financial institution via the network 102. The terminal device 5 is connected to the determination server 2 via the network 100 and the history server 3 via the network 102, and transmits / receives data to / from the determination server 2 and the history server 3.

  The PFP acquisition device 10 includes an insertion slot 4 into which the pattern matching card 6 is inserted, a camera 14 that acquires an image of the PFP registration unit 8 of the pattern matching card 6 inserted into the insertion slot 4, and acquisition by the camera 14. And a control device 12 that transmits the processed image data to the determination server 2. The PFP acquisition device 10 acquires an image of the PFP registration unit 8 of the pattern matching card 6 via the camera 14 and transmits this image data to the determination server 2. The PFP acquisition apparatus 10 is a device into which a credit card, a cash card, etc. including the PFP registration unit 8 are inserted.

  The camera 14 is a monochrome type CCD having effective pixels of 640 × 480 (about 300,000 pixels), for example. When the pattern matching card 6 is inserted into the insertion slot 4, the camera 14 has an area including the PFP registration unit 8 attached to the pattern matching card 6 (for example, 7.7 mm × A 5.7 mm area) is imaged by an optical system (lens) (not shown). Here, the observation area on the pattern matching card 6 per pixel is about 12 μm × 12 μm (7.7 mm / 640≈12 μm, 5.7 mm / 480≈12 μm). The camera 14 outputs the image data acquired by imaging the area including the PFP registration unit 8 to the control device 12 as an identification target pattern. The camera 14 acquires the PFP registration unit 8 on the pattern matching card 6 as image data of about 32 × 32 pixels to 64 × 64 pixels, for example. The camera 14 may be an image reading device such as a scanner.

  The history server 3 stores a usage history of the pattern matching card 6 and the like. This history includes the card user, use date and time, place of use, product information purchased using the card (product name, number of products, etc.), amount of money at that time, and the like. The history server 3 may be included in an external bank, securities company, credit card company, or the like. In this case, the determination server 2 may receive data such as usage history from the history server 3.

  The terminal device 5 accepts the type and amount of goods purchased using the pattern matching card 6 and transmits them to the determination server 2 and the history server 3. In this way, the terminal device 5 constitutes a reception unit that receives the number of products and the amount of money. The terminal device 5 receives data from the determination server 2, the history server 3, and the control device 12 and displays the data on a display device such as a liquid crystal display.

  The determination server 2 determines the identity between a pre-stored reference pattern composed of a random pattern (for example, a paper fiber pattern) on the object surface and image data (identification target pattern) transmitted by the PFP acquisition apparatus 10. To do. Further, the determination server 2 transmits the determination result to the history server 3, the terminal device 5, and the like. The determination server 2 will be described in detail later. The determination server 2 may be included in the control device 12.

FIG. 3 is a diagram illustrating a hardware configuration of the determination server 2, the history server 3, the terminal device 5, and the control device 12.
As illustrated in FIG. 3, the determination server 2 and the like include a CPU 18, a memory 20, a communication interface (IF) 22 that transmits and receives data to and from an external computer and camera 14 via a network, and a hard disk drive (HDD). A storage device 24 such as a CD / DVD device, a display device 26 such as a liquid crystal display, and an input device 28 including a keyboard / pointing device. These components are connected to each other by a bus 30. The determination server 2 or the like is, for example, a general-purpose computer in which a program described later is installed.

FIG. 4 is a diagram showing a functional configuration of the authenticity determination program 200 executed by the determination server 2.
As illustrated in FIG. 4, the authenticity determination program 200 includes a reference data storage unit 202, a correlation value calculation unit 204, a threshold control unit 206, and a comparison unit 208. The authenticity determination program 200 is loaded into the memory 20 and executed by the CPU 18. Note that all or some of the functions of the authenticity determination program 200 may be realized by hardware provided in the determination server 2.

  The determination server 2 stores a reference pattern composed of a random pattern on the surface of the object using these components, and the stored reference pattern and determination target pattern (that is, image data transmitted by the PFP acquisition apparatus 10). The correlation value is calculated, the threshold value is controlled, and the identity between the reference pattern and the determination target pattern is determined based on the calculated correlation value and the controlled threshold value.

  The reference data storage unit 202 stores a reference pattern including a random pattern attached to the PFP registration unit 8 of the pattern matching card 6 inserted in the PFP acquisition apparatus 10. The reference data storage unit 202 stores a plurality of reference patterns, and the plurality of reference patterns are uniquely identified by an identifier (ID). The reference pattern is, for example, image data of 32 × 32 pixels, and each pixel is represented by 256 gradations. The reference data storage unit 202 is realized by the memory 20, the storage device 24, and the like.

  The correlation value calculation unit 204 is a correlation between a reference pattern that is stored in the reference data storage unit 202 and has an identification target ID, and image data (determination target pattern) acquired by the PFP acquisition apparatus 10. Calculate the value. The ID is received by the terminal device 5 or the PFP acquisition device 10 and transmitted to the determination server 2.

More specifically, the correlation value calculation unit 204 sets a partial area in the input identification target pattern, sets the partial area image as a verification image, and calculates a correlation value between the verification image and the reference pattern. . In addition, the correlation value calculation unit 204 shifts the partial area by one pixel, calculates a correlation value between each of the plurality of verification images corresponding to these partial areas and the reference pattern, and obtains the maximum correlation value. . Correlation value calculation section 204 obtains normalized score N using the following equation.
N = (maximum correlation value−average correlation value) / standard deviation of correlation values (1)
The correlation value calculation unit 204 outputs the maximum correlation value and the normalized score to the comparison unit 208.

  The threshold control unit 206 controls the threshold and outputs the threshold to the comparison unit 208. Preferably, the threshold control unit 206 controls the threshold based on a determination history stored in advance. More specifically, the threshold control unit 206 inquires the history server 3 about the authenticity determination history at the time of card use stored in the history server 3 (FIG. 2), and obtains the authenticity determination history from the history server 3. If the percentage that is accepted and determined that the pattern matching card 6 is a fake is larger than a predetermined value, the threshold value is made larger than a preset value. The threshold value includes a first threshold value that is compared with the maximum correlation value by the comparison unit 208, which will be described later, and a second threshold value that is compared with the normalized score.

  The threshold control unit 206 may control the threshold based on the amount received by the terminal device 5 (FIG. 2). More specifically, the threshold control unit 206 accepts the amount of each of the commodities when using the card or the sum of these, and if the amount is greater than a predetermined value, the threshold control unit 206 increases the threshold above a preset value. .

Further, the threshold control unit 206 may control the threshold based on the number of products received by the terminal device 5. More specifically, the threshold control unit 206 receives the number of products purchased at one time using a card, and when the number of products is larger than a predetermined value, the threshold value is larger than a preset value. To do.
In addition, the threshold control unit 206 refers to the usage history, and sets a threshold in advance when the purchaser purchases a small amount of product (a product with an amount smaller than a predetermined value) at a frequency greater than a predetermined frequency. Make it larger than the specified value.

  The comparison unit 208 receives the threshold values (first threshold value and second threshold value) controlled by the threshold value control unit 206, and compares the maximum correlation value calculated by the correlation value calculation unit 204 with the first threshold value. Then, the calculated normalized score is compared with the second threshold value. When the comparison result satisfies the condition, the comparison unit 208 determines that the acquired identification target pattern and the reference pattern are the same (that is, true). For example, the comparison unit 208 determines to be true when the maximum correlation value is greater than the first threshold and the normalized score is greater than the second threshold. In this way, the comparison unit 208 constitutes a determination unit that determines the identity between the reference pattern and the determination target pattern based on the calculated correlation value and the controlled threshold value.

  In addition, since the width of the plant fiber forming the paper is 4 to 70 μm, if the resolution is 2400 dpi, the distribution of the feature amount derived from a very small number of plant fibers is reflected, and the determination accuracy of the identity is lowered. There is a case. Therefore, in order to obtain sufficient determination accuracy, the reference pattern image and the identification target pattern image are preferably regions of at least 16 × 16 pixels. Preferably, the reference pattern is 32 × 32 to 256 × 256 pixels, and the acquired identification target pattern is an area larger than the reference pattern. When the resolution of the camera 14 is high and the length of one side of the observation area per pixel is less than 4 μm, the determination server 2 performs sampling processing (mosaic processing or the like) on the received image data. After the application, the authenticity determination process may be performed.

FIG. 5 shows the similarity (correlation value, normalized score) of the identification target pattern with respect to the reference pattern when a pattern matching card having a different pattern is matched and when a pattern matching card having the same pattern as the reference pattern is matched. It is a figure which shows the relationship between and normalized frequency.
As shown in FIG. 5, when determining the authenticity of the pattern matching card 6 according to the threshold value Th, an error that determines the genuine pattern matching card 6 as a fake and the false pattern matching card 6 are determined as genuine. Errors can occur. The authenticity determination system 1 preferably sets the threshold Th so that both FRR and FAR are substantially zero. As described above, the authenticity determination system 1 according to the embodiment of the present invention controls the threshold Th based on the card user's usage history, usage amount, number of products, and the like.

FIG. 6 is a flowchart showing the authenticity determination program 200 authenticity determination processing (S10).
As shown in FIG. 6, in step 100 (S100), when the card user presents the pattern matching card 6 for product purchase, the store clerk or the like inputs the amount and quantity of the product to the terminal device 5 and presents it. The pattern matching card 6 thus inserted is inserted into the PFP acquisition device 10. The correlation value calculation unit 204 of the authenticity determination program 200 (FIG. 4) that operates on the determination server 2 includes the image data of the pattern verification card 6 acquired by the camera 14 of the PFP acquisition apparatus 10 (FIG. 2) and the pattern verification. It is determined whether or not the ID of the card 6 has been received from the control device 12. The authenticity determination program 200 proceeds to the process of S102 when the correlation value calculation unit 204 receives the identification target pattern, and returns to the process of S100 otherwise.

In step 102 (S102), the correlation value calculation unit 204 refers to the reference pattern corresponding to the ID stored in the reference data storage unit 202.
In step 104 (S104), the correlation value calculation unit 204 sets partial areas in the identification target pattern, shifts the partial areas one pixel at a time, and each of a plurality of matching images corresponding to these partial areas, and a reference The correlation value with the pattern is calculated, and the maximum correlation value is obtained. In addition, the correlation value calculation unit 204 obtains a normalized score N.

In step 106 (S106), the threshold value control unit 206 determines the authenticity determination history and usage history when using the card of the ID stored in the history server 3, and the amount and number of products transmitted from the terminal device 5. Based on this, the preset threshold values (first threshold value and second threshold value) are controlled.
In step 108 (S108), the comparison unit 208 compares the maximum correlation value calculated by the correlation value calculation unit 204 with the first threshold value controlled by the threshold value control unit 206, and the correlation value calculation unit 204 The calculated normalized score is compared with the second threshold value controlled by the threshold value control unit 206 to determine the authenticity of the presented pattern matching card 6.

  As described above, the authenticity determination system according to the present invention includes a storage unit that stores a reference pattern including a random pattern on an object surface, and a correlation value between the reference pattern stored in the storage unit and the determination target pattern. Based on the calculation means for calculating the threshold value, the threshold value control means for controlling the threshold value, the correlation value calculated by the calculation means and the threshold value controlled by the threshold value control means, the identity of the reference pattern and the determination target pattern is determined. Determination means for determining. Therefore, according to the authenticity determination system according to the present invention, the card user only presents the pattern matching card, and the threshold for authenticity determination is controlled, so that both convenience and safety are improved. Can do.

  In addition, since the authenticity determination system according to the present invention controls the threshold based on the determination history when using the card, the amount of money, and the number of products, the damage can be reduced and safety can be reduced without reducing convenience. Can be further enhanced.

  The authenticity determination system according to the present invention may be used for determining authenticity of a credit card, a cash card, or the like. The authenticity determination system according to the present invention can also be used for authenticity determination processing in an automatic teller machine (ATM).

  In addition, the authenticity determination system according to the present invention may control the use limit amount such as the purchase limit amount, the deductible limit amount and the like based on the determination history at the time of card use, the amount, the number of products, and the like. For example, a true / false determination system is purchased at a time when the rate at which a pattern matching card is determined to be a fake is greater than a predetermined value, or when the amount of each product when using the card or the sum of these is greater than a predetermined value. If the number of products to be purchased is larger than a predetermined value, or if the purchase is made by a purchaser who purchases a small amount of product (a product with an amount smaller than a predetermined value) at a frequency greater than a predetermined frequency, a usage limit is set in advance. Make it smaller than the existing value.

FIG. 1A illustrates a fiber pattern of paper, FIG. 1B illustrates a pattern verification card made of paper, and FIG. 1C illustrates plastic. Or the pattern collation card | curd which consists of a silver painting surface is shown. It is a figure which shows the authenticity determination system 1 which performs the authenticity determination process of the pattern collation card | curd 6. FIG. It is a figure which shows the hardware constitutions of the determination server 2, the log | history server 3, the terminal device 5, and the control apparatus 12. It is a figure which shows the function structure of the authenticity determination program 200 performed by the determination server 2. FIG. Similarity (correlation value, normalized score) of the identification target pattern with respect to the reference pattern when normalizing a pattern verification card having a different pattern, and when matching a pattern verification card having the same pattern as the reference pattern, and normalization It is a figure which shows the relationship with the frequency which performed. It is a flowchart which shows the authenticity determination program 200 authenticity determination processing (S10).

Explanation of symbols

1 authenticity determination system 2 determination server 3 history server 4 insertion slot 5 terminal device 6 pattern matching card 8 PFP registration unit 10 PFP acquisition device 12 control device 14 camera 18 CPU
20 Memory 22 Communication IF
24 storage device 200 authenticity determination program 202 reference data storage unit 204 correlation value calculation unit 206 threshold control unit 208 comparison unit

Claims (6)

Storage means for storing a reference pattern comprising a random pattern on the object surface;
Calculation means for calculating a correlation value between the reference pattern and the determination target pattern stored in the storage means;
Threshold control means for controlling the threshold;
A true / false determination system comprising: determination means for determining the identity of a reference pattern and a determination target pattern based on the correlation value calculated by the calculation means and the threshold value controlled by the threshold value control means. The authenticity determination system according to claim 1, wherein the threshold control unit controls the threshold based on a determination history stored in advance. It further has a receiving means for receiving the amount of money,
The authenticity determination system according to claim 1, wherein the threshold control unit controls the threshold based further on the amount received by the amount receiving unit. The receiving means further receives the number of products,
The authenticity determination system according to claim 3, wherein the threshold control unit controls the threshold based further on the number of products received by the receiving unit. Memorize a reference pattern consisting of a random pattern on the surface of the object,
Calculating a correlation value between the stored reference pattern and the determination target pattern;
Control the threshold,
A true / false determination method for determining identity between a reference pattern and a determination target pattern based on the calculated correlation value and the controlled threshold value. In a true / false judgment system including a computer,
A storage step for storing a reference pattern comprising a random pattern on the object surface;
A calculation step of calculating a correlation value between the stored reference pattern and the determination target pattern;
A threshold control step for controlling the threshold;
A program for causing a computer of the authenticity determination system to execute a determination step of determining the identity between a reference pattern and a determination target pattern based on the calculated correlation value and the controlled threshold value.

Images