Information processing device, method of execution by the information processing device, and program

Abstract

This technology provides a way to effectively detect impersonation using fraudulent data. [Solution] The information processing device 100 is connected to a network 111. The information processing device acquires authentication data, which is the biometric information of the person to be authenticated (authentication data acquisition unit 201), acquires fraudulent biometric data of the person by searching on the network (fraudulent data acquisition unit 202), and determines whether the authentication data is fraudulent or not based on a comparison of the authentication data and the fraudulent data (matching degree calculation unit 203, fraudulent data detection unit 204).

Description

Technical Field

[0001] This disclosure relates to an authentication technology for identifying a person based on biometric information.

Background Art

[0002] In an authentication technology system using biometric information such as a face, there is an act of impersonation to break through authentication by using attack materials such as a piece of paper printed with the face of a registered person or a tablet PC displaying the face (for example, Patent Documents 1 and 2).

[0003] Techniques for detecting such acts of impersonation are widely known. The technique of Patent Document 1 pre-registers known illegal data, and when the biometric information data of the authentication target matches the pre-registered illegal data, it determines that the biometric information of the identification target is known illegal data. Thereby, the act of impersonation is detected. Known illegal data is biometric information that has been illegally used in the past or may be illegally used.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, the technique of Patent Document 1 cannot detect an act of impersonation when unknown illegal data that has not been pre-registered is used for authentication. In recent years, it has become possible to easily obtain illegal data such as the face image of a registered person from the Internet or the like and attempt an act of impersonation. It is necessary to prevent such acts of impersonation using illegal data that can be easily obtained by a third party.

[0006] This disclosure provides technology for appropriately detecting impersonation activities using fraudulent data. [Means for solving the problem]

[0007] An information processing device according to one embodiment of the present disclosure is connected to a network. The information processing device includes a first acquisition means for acquiring authentication data, which is the biometric information of a person to be authenticated; a second acquisition means for acquiring fraudulent data of the biometric information of the person by searching on the network; and a determination means for determining whether the authentication data is fraudulent or not based on a comparison of the authentication data and the fraudulent data. [Effects of the Invention]

[0008] According to this disclosure, it is possible to appropriately detect impersonation activities using fraudulent data. [Brief explanation of the drawing]

[0009] [Figure 1] This is a block diagram showing an example of the hardware configuration of a computer device according to the first embodiment. [Figure 2] This is a block diagram showing an example configuration for the user authentication function. [Figure 3] This flowchart shows an example of the user authentication process in the first embodiment. [Figure 4] Figure 3 is a flowchart showing an example of the fraudulent data detection process in S303. [Figure 5] This block diagram shows an example of the configuration of the user authentication function in the second embodiment. [Figure 6] This flowchart shows an example of the user authentication process in the second embodiment. [Modes for carrying out the invention]

[0010] Embodiments of this disclosure will be described below with reference to the drawings. Not all of the features in the embodiments of this disclosure are essential, and features may be combined in any way. Furthermore, the configurations shown in the following embodiments are merely examples, and this disclosure is not limited to the illustrated configurations. In the drawings, the same or similar components are denoted by the same reference numerals to avoid redundant explanations.

[0011] [First Embodiment] Figure 1 is a block diagram showing an example of the hardware configuration of a computer device in one embodiment. The computer device 100 is an example of an information processing device.

[0012] The CPU 101 is a Central Processing Unit that controls the entire computer device 100. The ROM 102 is a Read Only Memory that stores programs and parameters that do not require modification. The RAM 103 is a Random Access Memory that temporarily stores programs and data supplied from the storage device 104 and other sources.

[0013] The storage device 104 includes a hard disk or memory card fixedly installed in the computer device 100. The storage device 104 may also include an external storage device. The external storage device is a storage medium such as a flexible disk (FD), a compact disk (CD) or other optical disk, a magnetic card, an optical card, an IC card, and / or other memory card that can be removed from the computer device 100.

[0014] The input device interface 105 is connected to the input device 109 and is an interface that receives input data from the input device 109. The input device 109 is a pointing device and / or keyboard that receives user input and inputs data.

[0015] The output device interface 106 is an interface that is connected to the monitor 110 and outputs image data to the monitor 110. The monitor 110 displays the data held by the computer device 100 and the image data supplied to the computer device 100.

[0016] The communication interface 107 is an interface for connecting to the network 111. The communication interface 107 is connected to the authentication data acquisition device 112 via the network 111.

[0017] The authentication data acquisition device 112 acquires authentication data. The authentication data is biometric information of a person to be authenticated as described later, and is typically image data including a face image. The authentication data acquisition device 112 is typically a camera (e.g., a network camera) that captures video. The authentication data acquisition device 112 includes one or more necessary sensors according to the type of biometric information to be acquired. For example, when the authentication data acquisition device 112 is a camera, in addition to an image sensor that receives visible light, it may include an image sensor that irradiates infrared light to the authentication target and receives the reflected light. The image sensor for visible light and the image sensor for infrared light may be configured as an integrated single image sensor, or may be separate image sensors. Alternatively, a plurality of authentication data acquisition devices 112 may be connected to the network 111, and each of the plurality of authentication data acquisition devices 112 may include sensors according to the type of biometric information to be acquired.

[0018] The authentication data acquisition device 112 may include a scanning mechanism using light or ultrasonic waves. As described later, when the biometric information to be acquired is fingerprint, iris, blood vessel, and / or voice, etc., the biometric information is read by the scanning mechanism. The above-described sensor or the scanning mechanism may include a microphone.

[0019] Network 111 includes, for example, at least one line of LAN and WAN, and includes one or more computer terminals, one or more servers, and / or a cloud computing system connected to the line. The line is composed of wired, wireless, or a combination thereof, and is mainly composed of the Internet.

[0020] System bus 108 is a transmission path that communicably connects CPU 101, ROM 102, RAM 103, storage device 104, input device interface 105, output device interface 106, and communication interface 107.

[0021] (Configuration of the personal authentication function) Figure 2 is a block diagram showing a configuration example of the personal authentication function. The functions of each of these blocks are realized by the CPU 101 executing a program stored in a readable storage medium such as the ROM 102.

[0022] Computer device 100 performs personal authentication processing by biometric authentication using the biometric information of the person to be authenticated. Hereinafter, the biometric information of the person to be authenticated is referred to as authentication data. In this embodiment, a facial image is given as a typical example of biometric information (authentication data), but it may be a partial image such as eyes, ears, and / or nose included in a person's head / face. Alternatively, the biometric information may be information regarding fingerprints, irises, blood vessels, and / or voices.

[0023] Computer device 100 uses the authentication data as a query to search for and acquire an image or video including the face of the person to be authenticated that corresponds to the illegal data on network 111. Illegal data is image data or video data including the face of the person to be authenticated that is uploaded on network 111 and can be easily obtained by a third party. Then, computer device 100 calculates the degree of coincidence between the authentication data and the illegal data, and if the degree of coincidence is greater than or equal to a certain value, it determines that the authentication data is illegal data.

[0024] The authentication data acquisition unit 201 acquires authentication data, which is a facial image of the person to be authenticated. The authentication data can be acquired, for example, via the authentication data acquisition device 112. In this embodiment, the hardware that implements the authentication data acquisition unit 201 mainly consists of the CPU 101 and the communication interface 107. The authentication data acquisition unit 201 is an example of a first acquisition means for acquiring authentication data, which is the biometric information of the person to be authenticated.

[0025] The fraudulent data acquisition unit 202 acquires fraudulent data. The fraudulent data acquisition unit 202 performs a reverse image search (RIS) on the network 111 using the facial image, which is authentication data, as a query. The fraudulent data acquisition unit 202 acquires images or videos containing the face of the person to be authenticated from the top N or more image data or video (video) data of the search results.

[0026] When the fraudulent data acquisition unit 202 acquires fraudulent data as video data, it divides the multiple frames that make up the video data (divides them into individual frames) and acquires them as a group of images containing multiple images. It is assumed that someone attempting to impersonate someone might use a video containing a face that has been searched on the network 111 for impersonation. In this case, in order to make the video data easier to use for impersonation, it is assumed that the face portion will be extracted from the video searched on the network 111, and an appropriate time portion will be extracted as a sequence of motion. By doing so, it is possible to create movements such as blinking and facial expressions on the face, making it more difficult for the data to be judged as non-biological, thus enabling more sophisticated impersonation. To counter this, this embodiment divides the video into frames and performs identity authentication processing (S306 in Figure 3, described later) on each of these still images (frames). The final identity authentication result is determined from the authentication results of each still image. As a method of determination, for example, a majority vote of the authentication results of each still image may be taken.

[0027] The details of the reverse image search described above will now be explained. In reverse image search, images belonging to the same category as the query image (authentication data) are retrieved. For example, if the query is an image of a person named A's face, various face images of A that belong to the same category will be retrieved. In this example, the same category is "person A" or "face images of person A". This search is performed using a machine learning model that has been trained to extract similar features from images of the same category and dissimilar features from images belonging to different categories. In the process of extracting similar features, a person's face is robust to face size, orientation, lighting variations, and changes over time. "Different categories" in the above example would be "other people" or "face images of other people" in relation to "person A" or "face images of person A".

[0028] The fraudulent data acquisition unit 202 is an example of a second acquisition means that acquires fraudulent data of the person's biometric information by searching on the network.

[0029] The matching degree calculation unit 203 calculates the degree of matching between the authentication data image acquired by the authentication data acquisition unit 201 and one or more images acquired as fraudulent data by the fraudulent data acquisition unit 202. The matching degree calculation unit 203 does not determine whether the faces of the people in the images match, but rather whether the image data itself matches.

[0030] However, even if a person attempting to impersonate someone uses an image included in the fraudulent data acquired by the fraudulent data acquisition unit 202 as authentication data, it is highly unlikely that the authentication data and the fraudulent data will be a perfect match. This is because geometric changes may occur when the authentication data acquisition device 112 captures the image, or the person attempting to impersonate someone may manipulate the image, potentially causing variations in the image. An example of how a person attempting to impersonate someone might manipulate an image is as follows: Cut out the paper with the face printed on it, following the outline of the face. • Shift the image display area so that the face shown on the device (smartphone or tablet computer) is in the center of the image.

[0031] Therefore, the degree of similarity is calculated on the premise that differences between images due to expected variations when images found through internet searches are displayed on a display medium are to be tolerated. This tolerance is called "variability tolerance." Display mediums include paper, screens, etc. Screens are not limited to monitor screens, but also include, for example, projector screens. When the display medium is paper, it is assumed that the image of the authentication data is printed on paper. Differences between images due to expected variations when images found through internet searches are displayed on a display medium are an example of a pre-predictable difference between the authentication data and the fraudulent data.

[0032] The following are examples of variations between images in terms of variability tolerance. • Geometric changes such as projection distortion • Changes in color tone • Image cropping • Noise addition

[0033] In this embodiment, the degree of matching is calculated by determining a hash value from the image, and a smaller Hamming distance between the hash values ​​of the images indicates a higher degree of matching. A perceptual hash is used as the hash value. To accommodate image cropping, the hash value may be calculated only from the face portion of the image.

[0034] The fraudulent data detection unit 204 determines that the authentication data acquired by the authentication data acquisition unit 201 is fraudulent data if the degree of matching calculated by the degree of matching calculation unit 203 exceeds a certain value (threshold).

[0035] The degree of agreement calculation unit 203 and the fraudulent data detection unit 204 are examples of determination means that determine whether or not the authentication data is fraudulent based on a comparison of the authentication data and the fraudulent data.

[0036] The non-biological data detection unit 205 determines whether the authentication data acquired by the authentication data acquisition unit 201 is biological or non-biological. Known methods are used for the determination. Known methods include, for example, a method of acquiring biological information such as body temperature, heart rate, and / or blood flow rate by irradiating the authentication target with infrared light. In this case, the non-biological data detection unit 205 makes a determination on the biological information acquired by the authentication data acquisition unit 201. The determination method may include machine learning or deep learning. The non-biological data detection unit 205 is an example of a biological determination means for determining whether or not the authentication data is non-biological.

[0037] The facial recognition unit 206 extracts features from the authentication data acquired by the authentication data acquisition unit 201 and compares them with the features of a pre-registered person's face, i.e., the face of the person in the identity authentication process, to perform identity authentication. The difference between the facial recognition unit 206 and the matching degree calculation unit 203 is that the matching degree calculation unit 203 calculates whether the image data itself is similar, while the facial recognition unit 206 determines whether the person whose face is included in the image matches the person in question. The facial recognition unit 206 is an example of an authentication means that performs biometric authentication on the target of authentication using the authentication data.

[0038] The data of a person's face image is stored in the storage device 104 or other device, and the face recognition unit 206 retrieves the registered person's face image from the storage device 104 or other device. The other device is, for example, a device connected to the computer device 100 via the network 111. The storage device 104 or other device may store in advance the person's face image and its corresponding feature quantities. The face recognition unit 206 may then retrieve the feature quantities corresponding to the person's face image in the personal authentication process described later.

[0039] In the facial recognition unit 206, a Convolutional Neural Network (CNN), a type of neural network, is used as the specific method for calculating features. A CNN is a method of extracting abstracted information from an input image by repeatedly performing a process consisting of convolution, activation, and pooling on the input image. The processing unit consisting of convolution, activation, and pooling is often called a hierarchy. Several well-known methods are known for the activation process, but one example is a method called Rectified Linear Unit (ReLU). Similarly, several well-known methods are known for the pooling process, but one example is a method called Max pooling. For example, as a CNN structure, one could use ResNet or similar, as described in non-patent literature (He, Kaiming, et al. “Identity mappings in deep residual networks.” Computer Vision-ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, October 11-14, 2016, Proceedings, Part IV 14. Springer International Publishing, 2016.).

[0040] (Another form of the calculation method by the degree of agreement calculation unit 203) Any method that can calculate the degree of similarity while allowing variation between two images is applicable, and is not limited to methods using the Hamming distance of hash values ​​between images. In another form, the degree of similarity calculation unit 203 may calculate the degree of similarity from the arrangement of multiple feature points extracted from the image. One example of a method for calculating the degree of similarity from the arrangement of feature points is a technique called Geometric Hashing.

[0041] In yet another configuration, the matching score calculation unit 203 may use a CNN-based calculation method, similar to the face recognition unit 206. The CNN used in the face recognition unit 206 aims to determine whether the faces in the images match. In contrast, the matching score calculation unit 203, which uses a CNN, extracts features to calculate whether the image data itself is similar, while allowing for variations.

[0042] In yet another configuration, the matching degree calculation unit 203 may calculate the matching degree of the image patterns by template matching.

[0043] (Identity verification process) Figure 3 is a flowchart showing an example of the user authentication process performed by the computer device 100. The user authentication process is achieved by the CPU 101 executing a program stored in a readable storage medium such as ROM 102.

[0044] In S301, the authentication data acquisition unit 201 acquires a facial image, which is authentication data. In S302, the fraudulent data acquisition unit 202 performs a network search using the facial image, which is authentication data acquired in S301, as a query.

[0045] In S303, the fraudulent data detection unit 204 performs fraudulent data detection processing. Details of the processing in S303 will be described later using the flowchart shown in Figure 4.

[0046] If fraudulent data is detected in S303, the user authentication process ends in S304; otherwise, the process moves to S305.

[0047] In S305, the non-biometric data detection unit 205 determines whether the authentication data is non-biometric data. If it is non-biometric data, the non-biometric data detection unit 205 determines that the authentication data is fraudulent data and terminates processing. Otherwise, processing proceeds to S306. As described above, a known method is used for the determination method in S305.

[0048] In S306, the facial recognition unit 206 determines, through facial recognition processing, whether the person whose face is included in the image of the authentication data matches a person who has been registered in advance.

[0049] (Detection of fraudulent data) Figure 4 is a flowchart showing an example of the fraudulent data detection process in S303 of Figure 3.

[0050] In S401, the fraudulent data detection unit 204 acquires images or videos containing the face of the person to be authenticated from the top N or more images or videos in the search results of S302 as fraudulent data. In addition, when the fraudulent data detection unit 204 acquires video data, it divides the video data frame by frame and acquires it as a group of images.

[0051] In S402, the fraudulent data detection unit 204 calculates a hash value from the authentication data obtained in S301. In this embodiment, a Perceptual Hash is used for the hash value.

[0052] S403 is the starting point of a loop of M images that have been acquired as fraudulent data and are subject to detection processing. If the fraudulent data is video data, M will be 2 or greater.

[0053] In S404, the fraudulent data detection unit 204 calculates a hash value from the fraudulent data. In this embodiment, a Perceptual Hash is used for the hash value.

[0054] In S405, the fraudulent data detection unit 204 calculates the Hamming distance between the hash values ​​of the authentication data and the fraudulent data. If the Hamming distance is below a certain value, the authentication data is determined to be fraudulent. In other words, if the degree of matching exceeds a threshold, the authentication data is determined to be fraudulent. If the fraudulent data is video data, the fraudulent data detection unit 204 only needs to determine that the authentication data is fraudulent if there is one or more frames where the degree of matching exceeds the threshold. S406 marks the end of the loop of M images that have been acquired as fraudulent data and are subject to detection processing.

[0055] (effect) As described above, in the user authentication process, the computer device 100 of this embodiment uses the authentication data as a query to search and obtain images or videos containing the face of the person to be authenticated that correspond to fraudulent data on the network 111. The computer device 100 then calculates the degree of matching between the authentication data and the fraudulent data, and if the degree of matching is above a certain value, it determines that the authentication data is fraudulent data. This makes it possible to appropriately detect impersonation activities using fraudulent data uploaded to the internet.

[0056] [Second Embodiment] In the first embodiment, the user authentication process performed a reverse image search on the network 111 (Internet) using the facial image, which is the authentication data, as the query. In the method of the first embodiment, data unrelated to the person being authenticated, such as the background in the image, is likely to be included in the search results. Therefore, in the user authentication process of the second embodiment, the name of the person being authenticated is used as the query to search for images. Below, we will mainly explain the parts that differ from the first embodiment, and omit the explanation of parts that are the same as the first embodiment.

[0057] (Configuration of the user authentication function) Figure 5 is a block diagram showing an example configuration of the user authentication function of this embodiment. The basic configuration in Figure 5 is the same as in Figure 2. The difference between Figure 2 and Figure 5 is that the user authentication function in Figure 5 includes a person name acquisition unit 502. Furthermore, the method of acquiring fraudulent data by the fraudulent data acquisition unit 503 differs from that of the first embodiment described above.

[0058] Computer device 100 performs identity verification using biometric authentication with the biometric information (authentication data) of the person to be authenticated. Computer device 100 uses the name of the person to be authenticated as a query to search and obtain images or videos containing the face of the person to be authenticated that correspond to fraudulent data on the network 111. Computer device 100 then calculates the degree of match between the authentication data and the fraudulent data, and if the degree of match is above a certain value, it determines that the authentication data is fraudulent data.

[0059] The person name acquisition unit 502 acquires the name of the person to be authenticated. For example, when the person to be authenticated uses the input device 109 to input their own name into the computer device 100, the person name acquisition unit 502 acquires the name of the person to be authenticated. Alternatively, the computer device 100 may pre-register pairs of people's facial images and names, and acquire the person's name through facial recognition. The method of acquiring a person's name is not limited to these.

[0060] The fraudulent data acquisition unit 503 acquires fraudulent data. The fraudulent data acquisition unit 503 uses the person's name acquired by the person's name acquisition unit 502 as a query to search for images on the network 111. As a result of the search, the fraudulent data acquisition unit 503 acquires images or videos that contain the face of the person to be authenticated from the top N or more images or videos. The fraudulent data acquisition unit 503 is an example of a third acquisition means for acquiring the name of the person to be authenticated.

[0061] The degree of agreement calculation unit 504 calculates the degree of agreement between the image of the authentication data acquired by the authentication data acquisition unit 501 and one or more images acquired as fraudulent data by the fraudulent data acquisition unit 503, in the same manner as in the first embodiment.

[0062] (Identity verification process) Figure 6 is a flowchart showing an example of the user authentication process performed by the computer device 100 in the second embodiment.

[0063] S601 and S604-S607 are the same as S301 and S303-S306 shown in Figure 3.

[0064] In S602, the person name acquisition unit 502 acquires the name of the person to be authenticated. In S603, the fraudulent data acquisition unit 503 performs a network search using the person name acquired in S602 as a query.

[0065] (effect) The user authentication process in the first embodiment described above performed a reverse image search on the network 111 using the facial image, which is the authentication data, as a query. In this search method, data unrelated to the person being authenticated, such as the background in the image, is likely to be included in the search results. In contrast, the user authentication process in this embodiment searches using the name of the person being authenticated as a query, thereby increasing the probability of obtaining data containing the face of the person being authenticated. As a result, more data containing the face of the person being authenticated can be obtained. This makes it possible to appropriately detect impersonation activities using fraudulent data uploaded to the internet.

[0066] [Other embodiments] This disclosure can also be implemented by supplying a program that implements one or more of the functions of the embodiments described above to a system or device via a network or storage medium, and by having one or more processors in the computer of that system or device read and execute the program. It can also be implemented by a circuit (e.g., an ASIC or FPGA) that implements one or more functions.

[0067] Although the present disclosure has been described in detail above based on its preferred embodiments, the present disclosure is not limited to the above embodiments, and various modifications are possible in accordance with the spirit of the present disclosure, and these modifications are not excluded from the scope of the present disclosure.

[0068] Embodiments in this disclosure include the following configurations, methods, and programs. (Composition 1) An information processing device connected to a network, A first acquisition means for acquiring authentication data, which is the biometric information of the person to be authenticated, A second acquisition means for obtaining unauthorized biometric data of the person through a search on the aforementioned network, The system includes a determination means for determining whether the authentication data is fraudulent based on a comparison of the authentication data and the fraudulent data. An information processing device characterized by the following: (Configuration 2) The system further includes a biometric determination means for determining whether the authentication data is non-biological or not. The information processing device according to configuration 1, characterized by the above. (Composition 3) The system further includes an authentication means that performs biometric authentication on the target of authentication using the aforementioned authentication data. An information processing device according to configuration 1 or 2, characterized by the above. (Composition 4) The biometric information of the aforementioned person is a facial image. An information processing device according to any one of configurations 1 to 3. (Composition 5) The determination means calculates the degree of agreement between the authentication data and the fraudulent data. An information processing apparatus according to any one of configurations 1 to 4, characterized by the above. (Composition 6) The determination means calculates the degree of agreement while allowing for pre-predictable differences between the authentication data and the fraudulent data. The information processing apparatus according to configuration 5, characterized by the features described herein. (Composition 7) If the acquired fraudulent data is video data, the determination means compares each of two or more frames from the multiple frames constituting the video data with the authentication data. An information processing device according to any one of configurations 1 to 6. (Composition 8) The system further includes a third means for obtaining the name of the person to be authenticated, The second acquisition means acquires fraudulent biometric data of a person corresponding to the name of the person acquired through a search on the network. An information processing device according to any one of configurations 1 to 7, characterized by the above. (method) A method performed by an information processing device connected to a network, The first acquisition step involves obtaining authentication data, which is the biometric information of the person to be authenticated. A second acquisition step involves obtaining fraudulent data of the person's biometric information through a search on the aforementioned network, The system includes a determination step of determining whether the authentication data is fraudulent based on a comparison of the authentication data and the fraudulent data. A method characterized by the following: (program) A program that causes the computer to execute each of the steps described in Configuration 9. [Explanation of symbols]

[0069] 111: Network 201, 501: Authentication data acquisition unit 202, 503: Unit for acquiring fraudulent data 203, 504: Matching degree calculation unit 204, 505: Fraudulent Data Detection Unit 205, 506: Non-biometric data detection unit 206, 507: Facial Recognition Section 502: Person name acquisition section

Claims

1. An information processing device connected to a network, A first acquisition means for acquiring authentication data, which is the biometric information of the person to be authenticated, A second acquisition means for obtaining unauthorized biometric data of the person through a search on the aforementioned network, The system includes a determination means for determining whether the authentication data is fraudulent based on a comparison of the authentication data and the fraudulent data. An information processing device characterized by the following:

2. The system further includes a biometric determination means for determining whether the authentication data is non-biological or not. The information processing apparatus according to feature 1.

3. The system further includes an authentication means that performs biometric authentication on the target of authentication using the aforementioned authentication data. The information processing apparatus according to feature 1.

4. The biometric information of the aforementioned person is a facial image. The information processing apparatus according to feature 1.

5. The determination means calculates the degree of agreement between the authentication data and the fraudulent data. The information processing apparatus according to feature 1.

6. The determination means calculates the degree of agreement while allowing for pre-predictable differences between the authentication data and the fraudulent data. The information processing apparatus according to feature 5.

7. If the acquired fraudulent data is video data, the determination means compares each of two or more frames from the multiple frames constituting the video data with the authentication data. The information processing apparatus according to feature 1.

8. The system further includes a third acquisition means for obtaining the name of the person to be authenticated, The second acquisition means acquires fraudulent biometric data of a person corresponding to the name of the person acquired through a search on the network. The information processing apparatus according to feature 1.

9. A method performed by an information processing device connected to a network, The first acquisition step involves obtaining authentication data, which is the biometric information of the person to be authenticated. A second acquisition step involves obtaining unauthorized biometric data of the person through a search on the aforementioned network, The system includes a determination step of determining whether the authentication data is fraudulent based on a comparison of the authentication data and the fraudulent data. A method characterized by the following:

10. A program that causes a computer to perform each of the steps described in claim 9.

Images