Information processing apparatus, information processing method, and recording medium

Abstract

This information processing apparatus is provided with a similarity index acquisition means that acquires similarity indexes, each indicating a similarity between two fingerprint images among a plurality of fingerprint images by changing a combination of the two fingerprint images, a coordinate allocation means that allocates two-dimensional coordinates to each of the plurality of fingerprint images such that a positional relationship corresponds to each of the similarity indexes, and an output means that arranges points corresponding to each of the plurality of fingerprint images at the allocated coordinates, and outputs two-dimensional distribution information representing, using a distribution of the points, the similarities between the plurality of fingerprint images.

Description

Information Processing Apparatus, Information Processing Method, and Recording Medium

[0001] This disclosure relates to the technical field of information processing apparatuses, information processing methods, and recording media.

[0002] Techniques for determining the similarity between pieces of information are known. For example, Patent Document 1 discloses a technique of obtaining a plurality of feature vectors (in a feature space) output from a learning model for inputs of a plurality of input images to a learned learning model, performing hierarchical clustering processing on the plurality of feature vectors, generating a plurality of hierarchical clusters, and extracting a subspace or vector corresponding to a specific cluster among the plurality of clusters as the concept (concept representation) of the specific cluster.

[0003] Japanese Patent Application Laid-Open No. 2023-136262

[0004] This disclosure aims to provide an information processing apparatus, an information processing method, and a recording medium for improving the techniques disclosed in prior art documents.

[0005] One aspect of an information processing apparatus includes a similarity index acquisition means for acquiring, by changing the combination of two fingerprint images among a plurality of fingerprint images, a similarity index indicating the similarity between the two fingerprint images; a coordinate assignment means for assigning two-dimensional coordinates to each of the plurality of fingerprint images so that the positional relationship corresponds to each of the similarity indices; and an output means for arranging points corresponding to each of the plurality of fingerprint images at the assigned coordinates and outputting two-dimensional distribution information representing the similarity between the plurality of fingerprint images by the distribution of the points.

[0006] One aspect of an information processing method includes acquiring, by changing the combination of two fingerprint images among a plurality of fingerprint images, a similarity index indicating the similarity between the two fingerprint images; assigning two-dimensional coordinates to each of the plurality of fingerprint images so that the positional relationship corresponds to each of the similarity indices; and arranging points corresponding to each of the plurality of fingerprint images at the assigned coordinates and outputting two-dimensional distribution information representing the similarity between the plurality of fingerprint images by the distribution of the points.

[0007] One embodiment of a recording medium contains a computer program that causes a computer to execute an information processing method which includes: acquiring similarity indices indicating the similarity between two fingerprint images from a plurality of fingerprint images, each with a different combination of the two fingerprint images; assigning two-dimensional coordinates to each of the plurality of fingerprint images so that their positional relationships correspond to each of the similarity indices; and outputting two-dimensional distribution information that represents the similarity between the plurality of fingerprint images as a distribution of the points, by placing points corresponding to each of the plurality of fingerprint images at the assigned coordinates.

[0008] This is a block diagram showing an example of the configuration of the information processing device related to this disclosure. This is a flowchart showing an example of the information processing operation of the information processing device related to this disclosure. This is a block diagram showing an example of the configuration of the information processing device related to this disclosure. This is a flowchart showing an example of the information processing operation of the information processing device related to this disclosure. This is an example of the output showing an example of the information processing operation of the information processing device related to this disclosure. This is a flowchart showing an example of the information processing operation of the information processing device related to this disclosure. This is a block diagram showing an example of the configuration of the information processing device related to this disclosure. This is an example of the output a block diagram showing an example of the configuration of the information processing device related to this disclosure.

[0009] The following describes embodiments of the information processing device, information processing method, and recording medium with reference to the drawings. [1: First Embodiment]

[0010] A first embodiment relating to an information processing device, an information processing method, and a recording medium will be described with reference to Figures 1 and 2. In the following description, the first embodiment relating to an information processing device, an information processing method, and a recording medium will be described using the information processing device 10.

[0011] As shown in Figure 1, the information processing device 10 includes a similarity index acquisition unit 11, a coordinate assignment unit 12, and a distribution information output unit 13. The similarity index acquisition unit 11 calculates an index indicating the similarity of two fingerprint images. The similarity index acquisition unit 11 may calculate the index indicating the similarity of two fingerprint images by a first method. The index indicating the similarity of two fingerprint images calculated by the first method is called the similarity index. The first method may include extracting features from the fingerprint images and comparing the extracted features with each other. The first method may include inputting the fingerprint images into a feature extraction model that extracts features and outputting the features.

[0012] The operations performed by the information processing device 10 will be explained with reference to the flowchart in Figure 2. As shown in Figure 2, the similarity index acquisition unit 11 calculates the similarity index of two fingerprint images from among a plurality of fingerprint images, changing the combination of the two fingerprint images (step S11). The coordinate assignment unit 12 assigns two-dimensional coordinates to each of the plurality of fingerprint images so that the positional relationship corresponds to each similarity index (step S12). The positional relationship refers to the distance between coordinates. For example, coordinates are assigned to each fingerprint image so that the distance between coordinates becomes the value of the similarity index of the fingerprint images corresponding to each coordinate.

[0013] The distribution information output unit 13 places points corresponding to each of the multiple fingerprint images in the assigned two-dimensional coordinates. The distribution information output unit 13 outputs two-dimensional distribution information that represents the similarity between the multiple fingerprint images as a distribution of points (step S13). The distribution information output unit 13 may, for example, display the distribution information. In other words, the distribution information output unit 13 outputs distribution information that represents the similarity between the multiple fingerprint images. In other words, the distribution information represents the similarity between fingerprint images based on the relative distance between points.

[0014] Thus, the information processing device 10 performs an information processing method that includes: obtaining a similarity index indicating the similarity between two fingerprint images from a plurality of fingerprint images, changing the combination of the two fingerprint images; assigning a two-dimensional coordinate to each of the plurality of fingerprint images so that the positional relationship corresponds to each similarity index; and outputting two-dimensional distribution information that represents the similarity between the plurality of fingerprint images as a distribution of points by placing the points corresponding to each of the plurality of fingerprint images at the assigned coordinates.

[0015] The information processing device 10 described above may be realized by a computer reading a computer program recorded on a recording medium. In this case, the computer program may cause the computer to execute an information processing method that includes: obtaining a similarity index indicating the similarity between two fingerprint images from a plurality of fingerprint images, changing the combination of the two fingerprint images; assigning two-dimensional coordinates to each of the plurality of fingerprint images so that the positional relationship corresponds to each similarity index; and outputting two-dimensional distribution information that represents the similarity between the plurality of fingerprint images as a distribution of points by placing points corresponding to each of the plurality of fingerprint images at the assigned coordinates. [Technical effects of the information processing device 10]

[0016] As described above, the information processing device 10 disclosed herein outputs distribution information representing the similarity between fingerprints. Since the distribution information representing the similarity between fingerprints is two-dimensional information, the information processing device 10 can make the similarity between fingerprints visible to humans. [2: Second Embodiment]

[0017] A second embodiment relating to an information processing device, an information processing method, and a recording medium will be described with reference to Figures 3 to 7. Hereinafter, the second embodiment relating to the information processing device, an information processing method, and a recording medium will be described using the information processing device 20. Note that, in the second embodiment, explanations that overlap with the description of the first embodiment described above will be omitted as appropriate. [2-1: Configuration of the Information Processing Device 20]

[0018] The configuration of the information processing device 20 will be explained with reference to Figure 3. Figure 3 is a block diagram showing the configuration of the information processing device 20.

[0019] As shown in Figure 3, the information processing device 20 comprises an arithmetic unit 21 and a storage device 22. Furthermore, the information processing device 20 may also include a communication device 23, an input device 24, and an output device 25. However, the information processing device 20 does not have to include the communication device 23 and at least one of the input device 24 and the output device 25. The arithmetic unit 21, the storage device 22, the communication device 23, the input device 24, and the output device 25 may be connected via a data bus 26.

[0020] The arithmetic unit 21 includes at least one processor (i.e., one or more processors) as hardware. The processor may include, for example, a processor conforming to a von Neumann computer architecture. A processor conforming to a von Neumann computer architecture may include at least one of a CPU (Central Processing Unit) and a GPU (Graphics Processing Unit). The processor may also include, for example, a processor conforming to a non-von Neumann computer architecture. A processor conforming to a non-von Neumann computer architecture may include at least one of an FPGA (Field Programmable Gate Array) and an ASIC (Application Specific Circuit).

[0021] The arithmetic unit 21 reads a computer program 221 which includes at least one of computer program code and computer program instructions. For example, the arithmetic unit 21 may read a computer program 221 stored in a storage device 22. For example, the arithmetic unit 21 may read a computer program 221 stored in a computer-readable and non-temporary recording medium using a recording medium reader (not shown) provided in the information processing device 20. The computer program 221 read from the recording medium may be stored in the storage device 22. The arithmetic unit 21 may obtain (i.e., download or read) a computer program 221 from a device (not shown) located outside the information processing device 20 via a communication device 23 (or other communication device). The downloaded computer program 221 may be stored in the storage device 22.

[0022] The arithmetic unit 21 executes the loaded computer program 221. As a result, logical functional blocks for executing the information processing that the information processing device 20 should perform are realized within the arithmetic unit 21. In other words, the arithmetic unit 21, together with the storage device 22 on which the computer program 221 is recorded (in other words, together with the storage device 22 and the computer program 221 recorded in the storage device 22), can function as a controller or computer for realizing logical functional blocks for executing the processing that the information processing device 20 should perform. That is, together with at least one processor in the arithmetic unit 21, the memory (recording medium) in the storage device 22 and the computer program 221 are configured so that the information processing device 20 performs the information processing that the information processing device 20 should perform.

[0023] Furthermore, the recording medium for recording the computer program 221 executed by the arithmetic unit 21 may include at least one of the following: optical discs such as CD-ROM, CD-R, CD-RW, flexible disk, MO, DVD-ROM, DVD-RAM, DVD-R, DVD+R, DVD-RW, DVD+RW, and Blu-ray (registered trademark); magnetic media such as magnetic tape; magneto-optical disks; semiconductor memory such as USB memory; and any other medium capable of storing a program. The recording medium may also include equipment capable of recording computer programs (for example, general-purpose or dedicated equipment on which the computer program 221 is implemented in an executable state in at least one form such as software and firmware). Furthermore, each process and function included in the computer program 221 may be realized by logical processing blocks implemented within the arithmetic unit 21 (i.e., the processor) when the arithmetic unit 21 executes the computer program 221, or by hardware such as a predetermined gate array (FPGA (Field Programmable Gate Array), ASIC (Application Specific Integrated Circuit)) provided by the arithmetic unit 21, or in a form in which logical processing blocks and partial hardware modules that realize some elements of the hardware are mixed.

[0024] The computing device 21 may implement a computational model that can be constructed by machine learning when the computing device executes a computer program 221. An example of a computational model that can be constructed by machine learning is a computational model that includes a neural network (so-called artificial intelligence (AI)). In this case, the learning of the computational model may include learning the parameters of the neural network (for example, at least one of the weights and biases). The computing device 21 may execute an information processing method using the computational model. In other words, the operation of executing the information processing method may include an operation using the computational model. Furthermore, the computing device 21 may implement a computational model that has been constructed by offline machine learning using training data. In addition, the computational model implemented in the computing device 21 may be updated by online machine learning on the computing device 21. Alternatively, the arithmetic unit 21 may perform speech processing using an arithmetic model implemented in an external device (i.e., a device located outside the information processing device 20) in addition to or instead of the arithmetic model implemented in the arithmetic unit 21.

[0025] Figure 3 shows an example of a logical functional block implemented in the arithmetic unit 21 for generating distribution information and performing fingerprint matching. As shown in Figure 3, the arithmetic unit 21 may include a similarity index acquisition unit 211, a coordinate assignment unit 212, a distribution information output unit 213, a division unit 214, a representative feature acquisition unit 215, and a matching unit 216. The "similarity index acquisition unit 211" is a component corresponding to the "similarity index acquisition unit 11" in the first embodiment described above, the "coordinate assignment unit 212" is a component corresponding to the "coordinate assignment unit 12" in the first embodiment described above, and the "distribution information output unit 213" is a component corresponding to the "distribution information output unit 13" in the first embodiment described above. The similarity index acquisition unit 211 may perform operations using a feature extraction model. The feature extraction model may be a specific example of the arithmetic model described above. The matching unit 216 may have a target reception unit 2161, a feature acquisition unit 2162, and a determination unit 2163.

[0026] The storage device 22 includes at least one memory capable of storing desired data. In other words, the storage device 22 includes at least one memory containing desired data. For example, the storage device 22 may store a computer program 221 executed by the arithmetic unit 21. In this case, the storage device 22 (memory) may be used as the recording medium described above for recording the computer program 221 executed by the arithmetic unit 21. The storage device 22 may temporarily store data that the arithmetic unit 21 temporarily uses when the arithmetic unit 21 is executing the computer program 221. The storage device 22 may store data that the information processing device 20 stores long-term. The storage device 22 may include at least one of RAM (Random Access Memory), ROM (Read Only Memory), hard disk drive, magneto-optical disk drive, SSD (Solid State Drive), and disk array drive. In other words, the storage device 22 may include a non-temporary recording medium.

[0027] The communication device 23 may be capable of communicating with devices outside the information processing device 20. The communication device 23 may use either wired or wireless communication.

[0028] The input device 24 is a device capable of receiving information input to the information processing device 20 from an external source. The input device 24 may include an operating device (e.g., a keyboard, mouse, touch panel, etc.) that can be operated by the user of the information processing device 20. The input device 24 may include a recording medium reader capable of reading information recorded on a recording medium that can be attached to and detached from the information processing device 20, such as a USB (Universal Serial Bus) memory. When information is input to the information processing device 20 via the communication device 23 (in other words, when the information processing device 20 acquires information via the communication device 23), the communication device 23 may function as an input device.

[0029] The output device 25 is a device capable of outputting information to the outside of the information processing device 20. The output device 25 may output visual information such as characters and images, auditory information such as sounds, or tactile information such as vibrations as the above information. The output device 25 may include, for example, at least one of a display, speaker, printer, and vibration motor. The output device 25 may also be capable of outputting information to a recording medium that can be attached to or detached from the information processing device 20, such as a USB memory. When the information processing device 20 outputs information via the communication device 23, the communication device 23 may function as an output device. [2-2: Feature point matching and AI matching]

[0030] Fingerprint matching may be performed by comparing feature information extracted from one fingerprint image with feature information extracted from another fingerprint image. For example, fingerprint matching may be performed by comparing feature points extracted from one fingerprint image with feature points extracted from another fingerprint image. Feature points include at least one of the endpoints and branching points of the fingerprint ridges. This fingerprint matching method is called "feature point matching." Feature point matching may determine the similarity of feature points extracted from fingerprint images and determine the similarity between fingerprints based on the similarity of the feature points. If the similarity between fingerprints is high, feature point matching may determine that the fingerprint images were taken from the same person.

[0031] One fingerprint image may be one from an unknown person. The other fingerprint image may be a registered fingerprint image. The registered fingerprint image may be one from a known person. The registered fingerprint image may be an image of a pressed fingerprint. The registered fingerprint image may be a pressed fingerprint taken for the purpose of registering it in a database, for example. The registered fingerprint image may be registered along with information indicating characteristic points.

[0032] Fingerprint matching may be performed by comparing features extracted from one fingerprint image with features extracted from another fingerprint image. The features may be information output by inputting the fingerprint image into a feature extraction model. The feature extraction model may be a model constructed to extract features. This method of fingerprint matching is called "AI matching". AI matching may determine the similarity of features extracted from fingerprint images and determine the similarity between fingerprints based on the similarity of the features. Similar to feature point matching, AI matching may determine that fingerprint images were taken from the same person if the similarity between fingerprints is high.

[0033] AI matching can process data much faster than feature point matching. Therefore, brute-force matching of registered fingerprint images is relatively easy with AI matching. On the other hand, while humans can visually determine whether feature points are similar or not, it is difficult for humans to understand the AI's recognition of similarity or dissimilarity in fingerprints. Specifically, it is difficult to visually perceive the basis for similarity determined by AI matching. For this reason, it is difficult to know whether the AI ​​is recognizing the similarity or dissimilarity of fingerprints as humans expect.

[0034] Therefore, in this embodiment, information that is understandable to humans is provided regarding the recognition of similarity and dissimilarity of fingerprints by AI. Furthermore, in this embodiment, the relatively fast AI matching process narrows down the matching targets for the relatively slow feature point matching process, performing only useful matches and reducing the processing load. [2-3: Information processing operations performed by the information processing device 20]

[0035] The information processing operations performed by the information processing device 20 will be explained with reference to Figure 4. Figure 4 is a flowchart showing an example of the flow of information processing operations performed by the information processing device 20. [2-3-1: Generation of distributed information]

[0036] As shown in Figure 4, the similarity index acquisition unit 211 acquires fingerprint images to be included in the distribution information (step S21). The similarity index acquisition unit 211 may acquire N registered fingerprint images as fingerprint images to be included in the distribution information. The following describes an example in which the information processing device 20 outputs distribution information for N registered fingerprint images.

[0037] The similarity index acquisition unit 211 may input fingerprint images into a feature extraction model and output features. The feature extraction model extracts features when fingerprint images are input. The similarity index acquisition unit 211 may input two fingerprint images into the feature extraction model and output two features. The similarity index acquisition unit 211 calculates the similarity between the two features as a similarity index. The method including feature extraction and similarity calculation by the similarity index acquisition unit 211 is called AI matching. The first method described above may be AI matching. The result of AI matching by the similarity index acquisition unit 211 is called the "AI matching score". The similarity index acquisition unit 211 acquires the AI ​​matching score (step S22).

[0038] The similarity index acquisition unit 211 may perform AI matching on any two of the N registered fingerprint images and output each AI matching score. Since there are N registered fingerprint images, there are N*(N-1) / 2 possible combinations of two registered fingerprint images. The similarity index acquisition unit 211 may perform N*(N-1) / 2 AI matching operations. In other words, the similarity index acquisition unit 211 may acquire N*(N-1) / 2 AI matching scores.

[0039] The AI ​​matching score may take values ​​between 0 and 1. In this case, the AI ​​matching score may be closer to 1, for example, as the registered fingerprint images are more similar to each other.

[0040] The coordinate assignment unit 212 assigns a two-dimensional coordinate to each of the N registered fingerprint images so that the positional relationship corresponds to each of the AI ​​matching scores (step S23). The coordinate assignment unit 212 may calculate a distance matrix based on the AI ​​matching scores. The coordinate assignment unit 212 may calculate the distance so that the more similar the registered fingerprint images are (the larger the AI ​​matching score), the closer they are. The coordinate assignment unit 212 may calculate the distance so that the less similar the registered fingerprint images are (the smaller the AI ​​matching score), the farther they are. If the AI ​​matching score can take values ​​from 0 to 1, the distance may be expressed as a value obtained by subtracting the AI ​​matching score from 1. The coordinate assignment unit 212 may obtain information indicating a distance of N*(N-1) / 2. The coordinate assignment unit 212 may obtain N two-dimensional coordinates based on the distance information.

[0041] The distribution information output unit 213 arranges points corresponding to each of the N registered fingerprint images at two-dimensional coordinates assigned thereto. The distribution information output unit 213 outputs two-dimensional distribution information representing the similarity between the N registered fingerprint images in the form of the distribution of points (step S24). The distribution information output unit 213 may, for example, control the output device 25 to output the distribution information. The distribution information output unit 213 may, for example, embed the registered fingerprint images in a two-dimensional space using Parametric UMA.

[0042] For example, FIG. 5 illustrates two-dimensional distribution information. It may also be said that FIG. 5 illustrates a two-dimensional space in which points representing the similarity between a plurality of fingerprint images are distributed. Each of the points distributed in the two-dimensional space corresponds to each of the registered fingerprint images. The similarity between the registered fingerprint images can be understood from the distance between the registered fingerprint images in the two-dimensional space. Each of the registered fingerprint images that are close in position in the two-dimensional space is similar to each other. Each of the registered fingerprint images that are far apart in position in the two-dimensional space is not similar to each other. For example, registered fingerprint images having the same pattern may be close in position in the two-dimensional space. Also, registered fingerprint images having different patterns may be far apart in position in the two-dimensional space. [2-3-2: Matching Based on Distribution Information] [a: Division of Distribution Information]

[0043] The division unit 214 divides the two-dimensional space including the coordinates into a plurality of sub-regions. The division unit 214 may, for example, divide the two-dimensional space including the coordinates so that the areas of the plurality of sub-regions are equal. The division unit 214 may, for example, divide the two-dimensional space including the coordinates by a grid as illustrated in FIG. 6.

[0044] The representative feature amount acquisition unit 215 may acquire a representative feature amount for each sub-region. The representative feature amount may be a feature amount representing a group of fingerprint images in which the coordinates are located within the range of the sub-region. The representative feature amount may be, for example, the average of the feature amounts, the feature amount of the registered fingerprint image located at the center of gravity of the sub-region, or the like. The average of the feature amounts may be obtained by a simple average or by a weighted average according to the distance.

[0045] The representative feature quantity acquisition unit 215 may acquire each of the representative feature quantities using a feature quantity extraction model. The representative feature quantity acquisition unit 215 may acquire representative feature quantities based on each of the feature quantities of the fingerprint image extracted by the feature quantity extraction model. [b: Determination of registered fingerprint images to be used in the second collation based on the first collation]

[0046] The first collation may be a collation by the first method described above, or may be the AI collation described above. The second collation may be a collation that takes longer to process and has a greater processing load than the first collation. On the other hand, the second collation may be a collation that uses fingerprint features that are easier for humans to understand than the first collation. For example, the second collation may be the feature point collation described above.

[0047] As shown in FIG. 7, the target reception unit 2161 receives an input of a target fingerprint image to be collated (step S25). The target fingerprint image may be associated with information indicating feature points included in the target fingerprint image. [First collation: Comparison of feature quantities extracted by a feature quantity extraction model]

[0048] The feature quantity acquisition unit 2162 acquires target feature quantities that are the feature quantities of the target fingerprint image (step S26). The feature quantity acquisition unit 2162 may acquire the target feature quantities using a feature quantity extraction model. The feature quantity acquisition unit 2162 may acquire the target feature quantities output by the feature quantity extraction model when the target fingerprint image is input.

[0049] The determination unit 2163 compares each of the target feature quantities with each of the representative feature quantities. The determination unit 2163 may determine representative feature quantities similar to the target feature quantities by comparing each of the target feature quantities with each of the representative feature quantities. The determination unit 2163 determines to collate each of the registered fingerprint images included in the fingerprint image group corresponding to the determined representative feature quantity with the target fingerprint image (step S27). That is, the determination unit 2163 may determine the registered fingerprint image to be collated with the target fingerprint image based on the distribution information. It may also be said that the determination unit 2163 determines which of at least any of the plurality of registered fingerprint images is to be collated with the target fingerprint image based on the distribution information.

[0050] The determination unit 2163 may determine at least one sub-region from a plurality of sub-regions. The determination unit 2163 may determine a sub-region corresponding to the representative feature most similar to the target feature. The determination unit 2163 may decide to compare each of the registered fingerprint images corresponding to the representative feature most similar to the target feature with the target fingerprint image. Alternatively, the determination unit 2163 may decide to perform a second comparison in order of how similar the representative features are to the target feature.

[0051] The matching unit 216 matches the target fingerprint image with at least one of the multiple registered fingerprint images (step S28). [Second matching: Comparison of feature points]

[0052] The matching unit 216 may perform a second matching between the target fingerprint image and each of the determined registered fingerprint images. The matching unit 216 may perform a feature point matching between the target fingerprint image and each of the determined registered fingerprint images. That is, in this embodiment, the AI ​​matching may identify a partial region containing a registered fingerprint area that is determined to correspond to the target fingerprint image, and perform a feature point matching with the registered fingerprint image contained in that partial region.

[0053] Alternatively, the matching unit 216 may control one or more matching mechanisms to match at least one of a plurality of registered fingerprint images with the target fingerprint image. For example, the matching unit 216 may control the execution of a second matching by a matching server located outside the information processing device 20. [2-4: Technical effects of the information processing device 20]

[0054] The information processing device 20 disclosed herein can provide human-understandable information regarding the recognition of similarity and dissimilarity of fingerprints by AI. Furthermore, the information processing device 20 can reduce the processing load by narrowing down the matching partners for feature point matching, which is processed relatively slowly, through relatively fast AI matching, and performing only useful matches. [3: Third Embodiment]

[0055] A third embodiment relating to an information processing device, an information processing method, and a recording medium will be described with reference to Figures 8 and 9. In the following description, the third embodiment relating to an information processing device, an information processing method, and a recording medium will be described using the information processing device 30. In the third embodiment, explanations that overlap with the descriptions of the first and second embodiments described above will be omitted as appropriate. In the drawings, parts common to the first and second embodiments will be denoted by the same reference numerals.

[0056] As shown in Figure 8, the arithmetic unit 21 of the information processing device 30 includes, as logical functional blocks, a similarity index acquisition unit 211, a coordinate assignment unit 212, a distribution information output unit 213, a division unit 314, a representative feature acquisition unit 215, and a matching unit 216. The third embodiment differs from the second embodiment in the division method for dividing the two-dimensional space including coordinates into a plurality of sub-regions.

[0057] In the third embodiment, the division unit 314 determines the size of the sub-regions such that the number of fingerprint images distributed in each sub-region is the same. Specifically, the division unit 314 determines the size of the sub-regions according to the density of fingerprint images distributed in the sub-regions. The division unit 314 may make the size of the sub-regions smaller when the density is high, and larger when the density is low. The size of the sub-regions is determined according to the density of fingerprint images distributed in the sub-regions. For example, as illustrated in Figure 9, if the shape of the sub-region is circular, the division unit 314 may make the diameter of circle 9a smaller when the density is high, and larger when the density is low. Also, if the shape of the sub-region is rectangular, the division unit 314 may make the side length shorter when the density is high, and longer when the density is low. Here, "density" may be, for example, the value obtained by dividing the amount of fingerprint image data corresponding to coordinates located within one sub-region by the area of ​​that sub-region. Alternatively, the density may be the value obtained by dividing the number of fingerprint image data corresponding to coordinates located within one sub-region by the area of ​​that sub-region. [Technical Effects]

[0058] The information processing device 30 disclosed herein can ensure that the fingerprint images distributed in each sub-region are the same. For example, when the determination unit 2163 decides to match each of the registered fingerprint images corresponding to the representative feature most similar to the target feature with the target fingerprint image, the same number of matches with the target fingerprint image can be performed regardless of which sub-region is determined. [4: Fourth Embodiment]

[0059] A fourth embodiment relating to an information processing device, an information processing method, and a recording medium will be described with reference to Figures 10 and 11. In the following description, the fourth embodiment relating to an information processing device, an information processing method, and a recording medium will be described using the information processing device 40. In the fourth embodiment, explanations that overlap with the descriptions of the first to third embodiments described above will be omitted as appropriate. In the drawings, parts common to the first to third embodiments will be denoted by the same reference numerals.

[0060] As shown in Figure 10, the arithmetic unit 21 of the information processing device 40 includes, as logical functional blocks, a similarity index acquisition unit 211, a coordinate assignment unit 212, a distribution information output unit 213, a division unit 414, a representative feature acquisition unit 215, and a matching unit 216. The fourth embodiment differs from the second and third embodiments in the division method for dividing the two-dimensional space including coordinates into a plurality of sub-regions.

[0061] In the fourth embodiment, the division unit 414 classifies multiple fingerprint images into multiple groups based on coordinates corresponding to the fingerprint images. For example, as illustrated in Figure 11, the division unit 414 may classify multiple registered fingerprint images into multiple groups according to the degree of dispersion of points included in the distribution information. The division unit 414 may define each of its sub-regions to include coordinates corresponding to fingerprint images belonging to the same group among the multiple groups.

[0062] The fingerprint images included in the partial region may have similar patterned designs. That is, the division section 414 may group fingerprint images having similar patterned designs.

[0063] The number of groups that the division section 414 classifies the data into can be determined arbitrarily. For example, the number of groups that the division section 414 classifies the data into can be determined according to the requirements for grouping fingerprint images. [Technical Effects]

[0064] The information processing device 40 disclosed herein can group registered fingerprint images according to their similarity. The information processing device 40 can, for example, perform a more detailed classification than existing pattern classifications. [5: Fifth Embodiment]

[0065] A fifth embodiment relating to an information processing device, an information processing method, and a recording medium will be described with reference to Figure 12. In the following description, the fifth embodiment relating to an information processing device, an information processing method, and a recording medium will be described using the information processing device 50. In the fifth embodiment, explanations that overlap with the descriptions of the first to fourth embodiments described above will be omitted as appropriate. In the drawings, parts common to the first to fourth embodiments will be denoted by the same reference numerals.

[0066] As shown in Figure 12, the arithmetic unit 21 of the information processing device 50 may include, as logical functional blocks, a similarity index acquisition unit 211, a coordinate assignment unit 212, a distribution information output unit 213, a division unit 214, a representative feature acquisition unit 215, and a matching unit 516. However, the information processing device 50 does not have to include the representative feature acquisition unit 215. Also, the matching unit 516 does not have to have a feature acquisition unit 2162. The fifth embodiment differs from the second to fourth embodiments in how the second matching is performed.

[0067] In the fifth embodiment, the matching unit 516 controls one or more matching mechanisms to compare at least one of the multiple fingerprint images with the target fingerprint image. The matching unit 516 may control multiple matching mechanisms to compare at least one of the multiple fingerprint images with the target fingerprint image. The matching unit 516 may also control the execution of a second matching by an external matching server of the information processing device 50.

[0068] In the fifth embodiment, the determination unit 5163 assigns at least one of a plurality of sub-regions to each of one or more matching mechanisms. If the information processing device 50 does not have a representative feature acquisition unit 215 and the matching unit 516 does not have a feature acquisition unit 2162, the determination unit 5163 may assign all of the plurality of fingerprint images to any of the plurality of matching mechanisms based on the plurality of sub-regions. In other words, the matching unit 516 may compare each of the plurality of fingerprint images with the target fingerprint image.

[0069] Furthermore, any of the division methods in the second embodiment, the third embodiment, and the fourth embodiment may be used for dividing the area into multiple sub-regions. Alternatively, a division method different from those in the second embodiment, the third embodiment, and the fourth embodiment may be used. [Technical Effects]

[0070] The information processing device 50 disclosed herein can distribute the processing load of the authentication process. In particular, when the data is divided into multiple sub-regions by the division method in the third embodiment, the amount of registered fingerprint image data processed by each of the multiple matching mechanisms becomes the same, so the processing load on each matching mechanism becomes the same. [6: Sixth Embodiment]

[0071] A sixth embodiment relating to an information processing device, an information processing method, and a recording medium will be described with reference to Figure 13. In the following description, the sixth embodiment relating to an information processing device, an information processing method, and a recording medium will be described using the information processing device 60. In the sixth embodiment, explanations that overlap with the descriptions of the first to fifth embodiments described above will be omitted as appropriate. In the drawings, parts common to the first to fifth embodiments will be denoted by the same reference numerals.

[0072] As shown in Figure 13, the arithmetic unit 21 of the information processing device 50 includes, as logical functional blocks, a similarity index acquisition unit 211, a coordinate assignment unit 212, a distribution information output unit 213, a division unit 214, a representative feature acquisition unit 215, a matching unit 216, and a learning unit 617.

[0073] The learning unit 617 acquires registered fingerprint images labeled with a correct class as training data. The correct class may include information indicating the correct answer for pattern classification. The learning unit 617 trains a feature extraction model so that the features of registered fingerprint images with the same correct label are similar. Alternatively, the learning unit 617 can be said to train the feature extraction model so that distribution information is generated in which points corresponding to registered fingerprint images with the same correct label are distributed close together. The learning unit 617 may train the parameters of the neural network included in the feature extraction model. The parameters of the neural network may be, for example, at least one of the weights and biases. [Technical Effects]

[0074] The information processing device 60 disclosed herein can provide a feature extraction model that extracts feature quantities from fingerprint images so that the feature quantities of fingerprint images that humans recognize as similar are similar. [7: Note]

[0075] Regarding the embodiments described above, the following additional notes are disclosed. [Addendum 1] An information processing apparatus comprising: a similarity index acquisition means for acquiring a similarity index indicating the similarity of two fingerprint images from a plurality of fingerprint images by changing the combination of the two fingerprint images; a coordinate assignment means for assigning two-dimensional coordinates to each of the plurality of fingerprint images so that the positional relationship corresponds to each of the similarity indexes; and an output means for outputting two-dimensional distribution information that represents the similarity between the plurality of fingerprint images as a distribution of the points, by arranging points corresponding to each of the plurality of fingerprint images at the assigned coordinates. [Addendum 2] The information processing apparatus according to Addendum 1, wherein the similarity index acquisition means inputs the two fingerprint images into a feature extraction model that extracts fingerprint features to output two features, and calculates the similarity between the two features as the similarity index. [Note 3] The information processing apparatus according to Note 2, comprising a matching means for matching the registered fingerprint image with the target fingerprint image when the fingerprint image is a registered fingerprint image and a different fingerprint image to be matched with the registered fingerprint image is a target fingerprint image, wherein the matching means determines the registered fingerprint image to be matched with the target fingerprint image based on the distribution information. [Note 4] The information processing apparatus according to Note 3, comprising a division means for dividing a two-dimensional space including the coordinates into a plurality of sub-regions. [Note 5] The information processing apparatus according to Note 4, further comprising a representative feature acquisition means for acquiring a representative feature, which is a feature representing the group of fingerprint images where the coordinates are located within the range of the sub-region, for each sub-region, wherein the matching means determines a representative feature similar to the feature of the target fingerprint image by comparing each of the acquired representative feature quantities with the feature quantities of the target fingerprint image, and decides to match each of the registered fingerprint images included in the group of fingerprint images corresponding to the determined representative feature quantity with the target fingerprint image. [Note 6] The information processing apparatus according to Note 4 or 5, wherein the dividing means determines the size of the partial region according to the density of the registered fingerprint images distributed in the partial region.[Note 7] The information processing apparatus according to Note 4 or 5, wherein the division means classifies the plurality of registered fingerprint images into a plurality of groups based on the coordinates corresponding to the registered fingerprint images, and defines each of the sub-regions to include the coordinates corresponding to the registered fingerprint images belonging to the same group among the plurality of groups. [Note 8] The information processing apparatus according to Note 4 or 5, wherein the matching means controls one or more matching mechanisms to match at least one of the plurality of registered fingerprint images with the target fingerprint image, and the matching means assigns at least one of the plurality of sub-regions to each of the matching mechanisms. [Note 9] The information processing apparatus according to Note 2, wherein fingerprint images labeled with a correct class are acquired as training data, and the feature extraction model is trained so that the feature quantities of fingerprint images with the same label are similar. [Note 10] A computer-based information processing method comprising: obtaining similarity indices indicating the similarity of two fingerprint images from a plurality of fingerprint images, each with a different combination of the two fingerprint images; assigning two-dimensional coordinates to each of the plurality of fingerprint images so that their positional relationships correspond to each of the aforementioned similarity indices; and outputting two-dimensional distribution information representing the similarity between the plurality of fingerprint images as a distribution of the points, by arranging points corresponding to each of the plurality of fingerprint images at the assigned coordinates. [Note 11] A recording medium on which a computer program is stored that causes a computer to execute an information processing device comprising: obtaining similarity indices indicating the similarity of two fingerprint images from a plurality of fingerprint images, each with a different combination of the two fingerprint images; assigning two-dimensional coordinates to each of the plurality of fingerprint images so that their positional relationships correspond to each of the aforementioned similarity indices; and outputting two-dimensional distribution information representing the similarity between the plurality of fingerprint images as a distribution of the points, by arranging points corresponding to each of the plurality of fingerprint images at the assigned coordinates.

[0076] Although this disclosure has been described above with reference to embodiments, this disclosure is not limited to the embodiments described above. Various modifications to the structure and details of this disclosure are possible, which can be understood by those skilled in the art within the scope of this disclosure. Furthermore, each embodiment can be combined with other embodiments as appropriate.

[0077] 10, 20, 30, 40, 50, 60 Information Processing Device 11, 211 Similarity Index Acquisition Unit 12, 212 Coordinate Assignment Unit 13, 213 Distribution Information Output Unit 214, 314, 414 Segmentation Unit 215 Representative Feature Acquisition Unit 216, 516 Matching Unit 2161 Target Reception Unit 2162 Feature Acquisition Unit 2163, 5163 Decision Unit 617 Learning Unit

Claims

1. An information processing device comprising: a similarity index acquisition means for acquiring a similarity index indicating the similarity between two fingerprint images from a plurality of fingerprint images, by changing the combination of the two fingerprint images; a coordinate assignment means for assigning two-dimensional coordinates to each of the plurality of fingerprint images so that their positional relationships correspond to each of the similarity indexes; and an output means for outputting two-dimensional distribution information that represents the similarity between the plurality of fingerprint images as a distribution of the points, by arranging points corresponding to each of the plurality of fingerprint images at the assigned coordinates.

2. The information processing device according to claim 1, wherein the similarity index acquisition means inputs the two fingerprint images into a feature extraction model that extracts fingerprint features to output two features, and calculates the similarity between the two features as the similarity index.

3. The information processing apparatus according to claim 2, wherein, when the fingerprint image is designated as a registered fingerprint image, and a different fingerprint image to be matched is designated as a target fingerprint image, the matching means for matching the registered fingerprint image with the target fingerprint image comprises the matching means for determining the registered fingerprint image to be matched with the target fingerprint image based on the distribution information.

4. The information processing apparatus according to claim 3, further comprising a division means for dividing a two-dimensional space including the coordinates into a plurality of sub-regions.

5. The information processing apparatus according to claim 4, further comprising a representative feature acquisition means for acquiring a representative feature, which is a representative feature of the group of fingerprint images whose coordinates are located within the range of the subregion, for each subregion, wherein the matching means determines a representative feature similar to the feature of the target fingerprint image by comparing each of the acquired representative feature quantities with the feature quantities of the target fingerprint image, and determines to match each of the registered fingerprint images included in the group of fingerprint images corresponding to the determined representative feature quantity with the target fingerprint image.

6. The information processing apparatus according to claim 4 or 5, wherein the dividing means determines the size of the partial region according to the density of the registered fingerprint images distributed in the partial region.

7. The information processing apparatus according to claim 4 or 5, wherein the matching means controls one or more matching mechanisms to compare at least one of the plurality of registered fingerprint images with the target fingerprint image, and the matching means assigns at least one of the plurality of partial regions to each of the matching mechanisms.

8. The information processing apparatus according to claim 2, which acquires fingerprint images labeled with a correct class as training data, and trains the feature extraction model so that the features of fingerprint images with the same label are similar.

9. A computer-based information processing method comprising: obtaining a similarity index indicating the similarity between two fingerprint images from a plurality of fingerprint images, for each of the two fingerprint images with a different combination; assigning a two-dimensional coordinate to each of the plurality of fingerprint images such that the positional relationship corresponds to each of the similarity indexes; and outputting two-dimensional distribution information that represents the similarity between the plurality of fingerprint images as a distribution of the points, by arranging the points corresponding to each of the plurality of fingerprint images at the assigned coordinates.

10. A recording medium on which a computer program is recorded that causes a computer to execute an information processing device, which includes: obtaining a similarity index indicating the similarity of two fingerprint images from a plurality of fingerprint images, for each of the two fingerprint images with a different combination; assigning a two-dimensional coordinate to each of the plurality of fingerprint images such that the positional relationship corresponds to each of the similarity indexes; and outputting two-dimensional distribution information that represents the similarity between the plurality of fingerprint images as a distribution of the points, by arranging a point corresponding to each of the plurality of fingerprint images at the assigned coordinate.

Images