Information processing device, information processing method, and recording medium

Abstract

This information processing device comprises: a probability estimation means for estimating, for each of a plurality of discrimination parts, a probability that a palmprint image includes a discrimination part; and a selection means for selecting, on the basis of the probability, the discrimination part included in the palmprint image from among the plurality of discrimination parts. The discrimination part is a part desired to be discriminated in a palm area and determined by dividing the palm area.

Description

Information processing device, information processing method, and recording medium

[0001] The present disclosure relates to the technical fields of an information processing device, an information processing method, and a recording medium.

[0002] It is known that processing to reduce the processing load of the matching operation is performed before matching palmprints. Patent Document 1 discloses that, for matching palmprints, features extracted from a palmprint image are used, and as preprocessing for feature extraction, the palmprint image is divided into a plurality of divided images, valid areas in the divided images are determined, processed divided images corresponding to the divided images are generated, and the processed divided images are integrated to generate a processed overall image corresponding to the palmprint image.

[0003] International Publication No. 2021 / 161375

[0004] This disclosure aims to improve upon the related art discussed above.

[0005] One aspect of the information processing device disclosed herein includes a probability estimation means for estimating the probability that a palm print image includes a distinct portion for each of a plurality of distinct portions, and a selection means for selecting, based on the probability, a distinct portion included in the palm print image from among the plurality of distinct portions, where the distinct portion is a portion of the palm region that is to be distinguished and is determined by dividing the palm region.

[0006] One aspect of the information processing method disclosed herein is an information processing method executed by a computer, which estimates the probability that a palm print image includes a distinct portion for each of a plurality of distinct portions, and selects a distinct portion included in the palm print image from among the plurality of distinct portions based on the probability, where the distinct portion is a portion of the palm region that is to be distinguished and is determined by dividing the palm region.

[0007] One aspect of the recording medium disclosed herein is an information processing method for estimating the probability that a palm print image includes a distinct portion for each of a plurality of the distinct portions, and selecting, based on the probability, a distinct portion included in the palm print image from among the plurality of the distinct portions, wherein the distinct portion is a portion of the palm region that is to be distinguished and is determined by dividing the palm region, and a computer program for causing a computer to execute the information processing method is recorded on the recording medium.

[0008] FIG. 1 is a block diagram showing the configuration of an information processing device according to the present disclosure. FIG. 2 is a flowchart showing the flow of information processing operation in the information processing device according to the present disclosure. FIG. 3 is a schematic diagram illustrating an overview of distinguishing portions. FIG. 4 is a block diagram showing the configuration of an information processing device according to the present disclosure. FIG. 5 is a flowchart showing the flow of information processing operation in the information processing device according to the present disclosure. FIG. 6 is a schematic diagram showing an information processing method in the information processing device according to the present disclosure. FIG. 7 is a block diagram showing the configuration of an information processing device according to the present disclosure. FIG. 8 is a flowchart showing the flow of information processing operation in the information processing device according to the present disclosure. FIG. 9 is a block diagram showing the configuration of an information processing device according to the present disclosure.

[0009] Hereinafter, an information processing device, an information processing method, and a recording medium according to an embodiment will be described with reference to the drawings. [1: First Embodiment]

[0010] A first embodiment of an information processing device, an information processing method, and a recording medium will be described below. Hereinafter, the first embodiment of the information processing device, the information processing method, and the recording medium will be described using an information processing device 1 according to this disclosure. [1-1: Configuration of Information Processing Device 1]

[0011] The configuration of an information processing device 1 according to this disclosure will be described with reference to Fig. 1. Fig. 1 is a block diagram showing the configuration of an information processing device 1 according to this disclosure.

[0012] 1, the information processing device 1 includes a calculation device 11, a storage device 12, and a communication device 13. The calculation device 11, the storage device 12, and the communication device 13 may be connected via a data bus 16.

[0013] The arithmetic device 11 includes at least one processor (i.e., one processor or multiple processors) as hardware. The processor may include, for example, a processor conforming to a von Neumann computer architecture. The processor conforming to the von Neumann computer architecture may include at least one of a CPU (Central Processing Unit) and a GPU (Graphics Processing Unit). The processor may include, for example, a processor conforming to a non-von Neumann computer architecture. The processor conforming to the non-von Neumann computer architecture may include at least one of an FPGA (Field Programmable Gate Array) and an ASIC (Application Specific Circuit).

[0014] The arithmetic device 11 reads a computer program 121 including at least one of computer program code and computer program instructions. For example, the arithmetic device 11 may read the computer program 121 stored in the storage device 12. For example, the arithmetic device 11 may read the computer program 121 stored in a computer-readable, non-transitory recording medium using a recording medium reading device (not shown) included in the information processing device 1. The computer program 121 read from the recording medium may be stored in the storage device 12. The arithmetic device 11 may acquire (i.e., download or read) the computer program 121 from a device (not shown) located outside the information processing device 1 via the communication device 13 (or another communication device). The downloaded computer program 121 may be stored in the storage device 12.

[0015] The arithmetic device 11 executes the loaded computer program 121. As a result, logical functional blocks for executing processing to be performed by the information processing device 1 (e.g., information processing described below) are realized within the arithmetic device 11. In other words, the arithmetic device 11, together with the storage device 12, etc. in which the computer program 121 is recorded (in other words, together with the storage device 12 and the computer program 121 recorded in the storage device 12, etc.), can function as a controller or computer for realizing the logical functional blocks for executing processing to be performed by the information processing device 1. In other words, the at least one processor included in the arithmetic device 11, the memory (recording medium) included in the storage device 12, etc., and the computer program 121 are configured to cause the information processing device 1 to perform processing to be performed by the information processing device 1 (e.g., information processing described below). The arithmetic device 11 may output information to another computer, cloud server, or other device (not shown) provided outside the information processing device 1 via the communication device 13 (or other communication device).

[0016] The recording medium for recording the computer program 121 executed by the arithmetic device 11 may be at least one of a 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) optical disk, magnetic medium such as magnetic tape, magneto-optical disk, semiconductor memory such as USB memory, and any other medium capable of storing a program. The recording medium may include a device capable of recording the computer program 121 (for example, a general-purpose device or a dedicated device in which the computer program 121 is implemented in a state in which it can be executed in at least one of the forms of software and firmware). Furthermore, each process or function included in the computer program 121 may be realized by a logical processing block realized within the arithmetic device 11 when the arithmetic device 11 (i.e., processor) executes the computer program 121, or may be realized by hardware such as a predetermined gate array (FPGA (Field Programmable Gate Array), ASIC (Application Specific Integrated Circuit)) included in the arithmetic device 11, or may be realized in a form that mixes logical processing blocks and partial hardware modules that realize some elements of the hardware.

[0017] A computation model M that can be constructed by machine learning is implemented in the computation device 11 by the computation device 11 executing the computer program 121. An example of the computation model M that can be constructed by machine learning is a computation model M including a neural network (so-called artificial intelligence (AI)). In this case, learning of the computation model M may include learning of parameters of the neural network (for example, at least one of a weight and a bias). The computation device 11 executes at least an estimation process using the computation model M. The computation device 11 may be implemented with a computation model M that has been constructed by machine learning. The computation device 11 may be implemented with a computation model M that has been constructed by offline machine learning using teacher data. Furthermore, the computation model M implemented in the computation device 11 may be updated by online machine learning on the computation device 11. Alternatively, the calculation device 11 may perform information processing using a calculation model M implemented in a device external to the calculation device 11 (i.e., a device provided outside the information processing device 1) in addition to or instead of the calculation model M implemented in the calculation device 11.

[0018] 1 shows an example of logical functional blocks implemented in a computing device 11 for executing information processing. As shown in FIG. 1, an estimation unit 111 and a selection unit 112 are implemented in the computing device 11. The estimation unit 111 executes inference processing using the above-described computation model M. The processing performed by the estimation unit 111 and the selection unit 112 will be described with reference to FIG. 2.

[0019] The storage device 12 includes at least one memory capable of storing desired data. In other words, the storage device 12 includes at least one memory containing desired data. For example, the storage device 12 may store a computer program 121 executed by the arithmetic device 11. In this case, the storage device 12 (memory) may be used as the above-mentioned recording medium for recording the computer program 121 executed by the arithmetic device 11. The storage device 12 may temporarily store data used by the arithmetic device 11 when the arithmetic device 11 is executing the computer program 121. The storage device 12 may also store data to be stored long-term by the information processing device 1. The storage device 12 may include at least one of a RAM (Random Access Memory), a ROM (Read Only Memory), a hard disk device, a magneto-optical disk device, an SSD (Solid State Drive), and a disk array device. In other words, the storage device 12 may include a non-temporary recording medium.

[0020] The communication device 13 can communicate with devices external to the information processing device 1 or 2 via a communication network (not shown). The communication device 13 may be a communication interface based on standards such as Ethernet (registered trademark), Wi-Fi (registered trademark), Bluetooth (registered trademark), or USB (Universal Serial Bus). [1-2: Information Processing Method Executed by the Information Processing Device 1]

[0021] An information processing method executed by the information processing device 1 will be described with reference to Fig. 2. Fig. 2 is a flowchart showing an example of the flow of the information processing method executed by the information processing device 1.

[0022] As shown in Fig. 2, the estimation unit 111 estimates the probability that the palm print image contains distinct regions (step S11). Distinct regions are regions in the palm region that are determined by dividing the palm region and that are to be distinguished. The estimation unit 111 performs the estimation using a computational model M. When a palm print image is input, the computational model M outputs the probability that the palm print image contains each of the multiple distinct regions. The estimation unit 111 estimates the probability that each of the multiple distinct regions contains the distinct region.

[0023] The selection unit 112 selects a distinctive portion included in the palm print image from among the plurality of distinctive portions based on the probability that the distinctive portion is included (step S12). [1-3: Technical Effects of Information Processing Device 1]

[0024] The information processing device 1 according to this disclosure can select a distinguishing region to which the palm print image corresponds using the calculation model M. [2: Second embodiment]

[0025] A second embodiment of an information processing device, an information processing method, and a recording medium will be described below. Hereinafter, the second embodiment of an information processing device, an information processing method, and a recording medium will be described using an information processing device 2 according to this disclosure. [2-1: Distinguishing Part]

[0026] In this embodiment, the portion of the palm region that is to be distinguished and defined by dividing the palm region is referred to as a "distinguishing portion." The "palm region" in this embodiment may include the palm (palm) as shown in FIG. 3( a) and the side of the palm on the little finger side as shown in FIG. 3( b). Furthermore, the "palm print" in this embodiment may include the pattern on the palm (palm) and the pattern on the side of the palm on the little finger side.

[0027] As illustrated in FIG. 3( a), the distinguishing regions may be defined by dividing the palm, for example, by a first line L1 connecting the metacarpophalangeal joints of the hand and a second line L2 extending from the first line L1. The second line L2 may be perpendicular to the first line L1. The second line L2 may extend from the center of the first line L1. In this case, the palm area of ​​one hand can be divided into distinguishing regions, including the interdigital region I (base of the fingers), the thenar region T (thenar), and the hypothenar region H (hypothenar) illustrated in FIG. 3( a), and the lateral palm region W (whiter's) illustrated in FIG. 3( b). The first line L1 may be rephrased as a line that distinguishes the finger base region I from the hypothenar region H and / or the thenar region. In other words, the second line L2 may be a line that distinguishes the little toe ball area H from the thumb ball area.

[0028] The base of the fingers I is the base of the four fingers other than the thumb on the palm. If the first line is a line connecting the metacarpophalangeal joints of the hand, the base of the four fingers I is the area on the side of the four fingers other than the thumb from the metacarpophalangeal joints of the hand. The thenar eminence T is the area on the thumb side of the palm on the wrist side. If the first line is a line connecting the metacarpophalangeal joints of the hand, the thenar eminence T is the area on the thumb side of the palm on the wrist side from the metacarpophalangeal joints of the hand. The hypothenar eminence H is the area on the little finger side of the palm on the wrist side. If the first line is a line connecting the metacarpophalangeal joints of the hand, the hypothenar eminence H is the area on the little finger side of the palm on the wrist side from the metacarpophalangeal joints of the hand. The lateral palm area (W: Witter's) is the side of the palm on the little finger side.

[0029] In this case, the palm regions of both hands include the finger base region RI, the thenar region RT, the hypothenar region RH, and the palmar region RW of the right hand, as well as the finger base region LI, thenar region LT, the hypothenar region LH, and the palmar region LW of the left hand.

[0030] A palm print image may include up to three distinct regions. Specifically, the palm print image may include the base of the fingers RI of the right hand, the thumb ball RT of the right hand, and the hypothenar region RH of the right hand. The palm print image may also include the lateral palm region RW of the right hand, the base of the fingers RI of the right hand, and the hypothenar region RH of the right hand. Similarly, the palm print image may include the base of the fingers LI of the left hand, the thumb ball LT of the left hand, and the hypothenar region LH of the left hand. The palm print image may also include the lateral palm region LW of the left hand, the base of the fingers LI of the left hand, and the hypothenar region LH of the left hand.

[0031] A palm print image may include two adjacent distinguishing regions. Alternatively, a palm print image may include a single distinguishing region. [2-2: Configuration of Information Processing Device 2]

[0032] The configuration of the information processing device 2 according to this disclosure will be described with reference to Fig. 4. Fig. 4 is a block diagram showing the configuration of the information processing device 2 according to this disclosure.

[0033] 4, the information processing device 2 may further include an input device 14 and an output device 15 in addition to the calculation device 11, the storage device 12, and the communication device 13. However, the information processing device 2 does not necessarily have to include at least one of the input device 14 and the output device 15. The calculation device 11, the storage device 12, the communication device 13, the input device 14, and the output device 15 may be connected via a data bus 16.

[0034] 4, in addition to an estimation unit 211 and a selection unit 212, an acquisition unit 213 and a matching unit 214 are further implemented in the calculation device 11 in the second embodiment. Furthermore, a registered image database DB may be implemented in the storage device 12 in the second embodiment. Note that the registered image database DB may be implemented in a storage device outside the information processing device 2.

[0035] The input device 14 is a device that accepts information input to the information processing device 2 from outside the information processing device 2. For example, the input device 14 may include an operation device (e.g., at least one of a keyboard, a mouse, and a touch panel) that can be operated by an operator of the information processing device 2. For example, the input device 14 may include a reading device that can read information recorded as data on a recording medium that can be externally attached to the information processing device 2.

[0036] The output device 15 is a device that outputs information to the outside of the information processing device 2. For example, the output device 15 may output information as an image. That is, the output device 15 may include a display device (a so-called display) that can display an image showing the information to be output. For example, the output device 15 may output information as sound. That is, the output device 15 may include an audio device (a so-called speaker) that can output sound. For example, the output device 15 may output information on paper. That is, the output device 15 may include a printing device (a so-called printer) that can print desired information on paper.

[0037] The information processing device 2 is configured as a device for selecting a distinctive part, similar to the information processing device 1. Furthermore, the information processing device 2 is configured as a device for matching palm prints. The information processing device 2 matches palm prints based on the selected distinctive part. [2-3: Information Processing Method Executed by the Information Processing Device 2]

[0038] The information processing operation in the information processing device 2 will be described with reference to Fig. 5. Fig. 5 is a flowchart showing the flow of the information processing operation in the information processing device 2.

[0039] As shown in Fig. 5, upon receiving a palm print image matching request, the acquisition unit 213 acquires a palm print image (step S21). The palm print image matching request is a request to match the palm print image with a registered image registered in the registered image database DB. In this embodiment, the palm print image may be an image of the entire palm print or an image of a portion of the palm print. For example, the palm print image may be an image of a leftover palm print. The palm print image may also be a rectangular image.

[0040] The registered images registered in the registered image database DB are images of distinctive regions. When the distinctive regions are defined as shown in Fig. 3, the registered images may include an image of the finger base region RI of the right hand, an image of the thenar region RT of the right hand, an image of the hypothenar region RH of the right hand, an image of the palmar region RW of the right hand, an image of the finger base region LI of the left hand, an image of the thenar region LT of the left hand, an image of the hypothenar region LH of the left hand, and an image of the palmar region LW of the left hand.

[0041] The registered image database DB may be constructed for each distinct body part. That is, a DB for the finger base part LI of the left hand may be constructed. Alternatively, the registered images registered in the registered image database DB may be accompanied by information that enables identification of which distinct body part the image belongs to.

[0042] Note that, for a certain individual, images of the eight types of distinctive parts exemplified in FIG. 3 may be registered in the registered image database DB. Furthermore, for another individual, images of any of the eight types of distinctive parts exemplified in FIG. 3 may be registered. In other words, images of all eight types of distinctive parts exemplified in FIG. 3 may not necessarily be registered in the registered image database DB for a certain individual. Note that the registered images registered in the registered image database DB do not have to be images of each distinct part. The registered image registered in the registered image database DB may be an image of the entire palm print.

[0043] The estimation unit 211 estimates the probability that the palm print image includes a distinctive portion using the computation model M (step S22). The estimation unit 211 estimates the probability that the distinctive portion is included for each of a plurality of distinctive portions.

[0044] The computational model M is constructed by machine learning. The computational model M may be constructed by machine learning using training data. The training data may be data including a training palm print image and correct answer information indicating distinctive parts contained in the training palm print image. The computational model M may be trained so that, when a training palm print image is input, it highly likely infers the distinctive parts contained in the training palm print image. The training of the computational model M will be described further in other embodiments.

[0045] As illustrated in FIG. 6 , when a palmprint image I is input, the computational model M outputs the probability that the palmprint image includes each of a plurality of distinct regions (base of the fingers RI of the right hand, ball of the thumb RT of the right hand, ball of the hypothenar region RH of the right hand, palm lateral region RW of the right hand, base of the fingers LI of the left hand, ball of the thumb LT of the left hand, ball of the hypothenar region LH of the left hand, and palm lateral region LW of the left hand). FIG. 6 illustrates an example in which the computation model M has inferred that the probability that palmprint image I includes the finger base region RI of the right hand is "0.00", the probability that the palmprint image includes the thenar region RT of the right hand is "0.00", the probability that the palmprint image includes the hypothenar region RH of the right hand is "0.00", the probability that the palmprint image includes the lateral palm region RW of the right hand is "0.00", the probability that the palmprint image includes the finger base region LI of the left hand is "0.01", the probability that the palmprint image includes the thenar region LT of the left hand is "0.95", the probability that the palmprint image includes the hypothenar region LH of the left hand is "0.99", and the probability that the palmprint image includes the lateral palm region LW of the left hand is "0.06".

[0046] The selection unit 212 selects a distinctive portion included in the palm print image from among the plurality of distinctive portions based on the probability that the distinctive portion is included (step S23). The selection unit 212 may select a distinctive portion for which the probability of including the distinctive portion exceeds a predetermined value. For example, the predetermined value may be "0.9". In the example shown in FIG. 6, the selection unit 212 may select "LT" and "LH". In other words, the selection unit 212 may select a plurality of distinctive portions.

[0047] The matching unit 214 matches the registered image of the distinctive part selected by the selection unit 212 with the palm print image (step S24). If the registered image is an image of the entire palm print, the matching unit 214 may cut out the area of ​​the relevant distinctive part from the entire palm print and match it with the palm print image.

[0048] That is, the selection unit 212 selects distinct parts to be matched by the matching unit 214. The matching unit 214 may extract feature amounts from each of the selected distinct part registration image and the palm print image, and match the selected distinct part registration image with the palm print image based on the similarity of the feature amounts.

[0049] Limiting the objects to be matched In other words, the selection unit 212 excludes, from the plurality of distinctive portions, those that do not need to be objects to be matched. In other words, the selection unit 212 leaves, from the plurality of distinctive portions, those that need to be objects to be matched.

[0050] That is, the selection unit 212 limits the objects to be matched, thereby reducing the number of objects to be matched. Therefore, this embodiment can be applied to pre-processing of latent palm print matching. [2-3: Technical Effects of Information Processing Device 2]

[0051] For example, when matching partial palm prints such as latent palm prints, matching time can be reduced by limiting the areas to be matched. The identification of the areas to be matched is often determined based on the experience of the forensic examiner.

[0052] The information processing device 2 according to this disclosure uses artificial intelligence to estimate the corresponding body part, thereby realizing high-speed, high-accuracy, and stable body part identification without relying on experience. Estimation using artificial intelligence allows for stable, accurate estimation without relying on experience, and enables faster estimation compared to estimation performed manually. [3: Third embodiment]

[0053] A third embodiment of an information processing device, an information processing method, and a recording medium will be described below. Hereinafter, the third embodiment of an information processing device, an information processing method, and a recording medium will be described using an information processing device 3 according to this disclosure. [3-1: Configuration of Information Processing Device 3]

[0054] Like the information processing device 1 and the information processing device 2, the information processing device 3 is configured as a device for selecting distinctive parts. Also, like the information processing device 2, the information processing device 3 is configured as a device for matching palm prints. Furthermore, the information processing device 3 is configured as a device for estimating the orientation of a palm print image. The information processing device 3 matches palm prints based on the selected distinctive parts and the inclination of the palm print image.

[0055] 7 , the calculation device 11 according to the third embodiment further includes a rotation control unit 316 in addition to an estimation unit 311, a selection unit 212, an acquisition unit 213, and a matching unit 214. The estimation unit 311 according to the third embodiment includes a probability estimation unit 3111 and a tilt estimation unit 3112.

[0056] The tilt estimation unit 3112 estimates the tilt of the palm print image with respect to a predetermined direction using the calculation model M. The predetermined direction may be the forward direction of the palm print. The direction of the line perpendicular to the first line is referred to as the forward direction of the palm print.

[0057] In the third embodiment, when a palm print image is input, the computation model M may output information indicating the tilt of the palm print image in addition to the probability that the palm print image includes a distinctive part. Alternatively, the selection unit 312 may use a computation model M that outputs the probability that the palm print image includes a distinctive part, and a computation model M that outputs information indicating the tilt of the palm print image. In other words, there may be a computation model for part inference and a computation model for tilt inference. In this case, a palm print image is input to each computation model M. [3-2: Information Processing Method Executed by Information Processing Device 3]

[0058] As shown in FIG. 8, upon receiving a request to match a palm print image, the acquisition unit 213 acquires a palm print image (step S21).

[0059] The probability estimation unit 3111 estimates the probability that the palm print image contains a distinctive portion using the computational model M (step S22). The selection unit 212 selects a distinctive portion contained in the palm print image from among the multiple distinctive portions, based on the probability that the distinctive portion is contained (step S23).

[0060] The tilt estimation unit 3112 uses the calculation model M to estimate the tilt of the palm print image with respect to the forward direction of the palm print (step S31). For example, in the example shown in FIG. 9, direction d1 may be the direction of palm print image I. Also, direction d2 may be the forward direction of the palm print. In the example shown in FIG. 9, the tilt estimation unit 3112 may calculate the difference between direction d1 and direction d2 and estimate the tilt of direction d1 with respect to direction d2. The tilt estimation unit 3112 may estimate, for example, that the image is tilted 10 degrees counterclockwise.

[0061] The rotation control unit 316 limits the range of rotation of the palm print image and the registered image based on the tilt (step S32). The rotation control unit 316 may rotate the palm print image so that direction d2 faces directly upward. Alternatively, the rotation control unit 316 may rotate the registered image so that the forward direction of the palm print in the registered image faces direction d2. If the palm print image is estimated to be tilted -10 degrees with respect to the forward direction of the palm print as described above, the rotation control unit 316 may control at least one of the palm print image and the registered image to rotate in 1-degree increments, for example, within a range from -20 degrees to 0 degrees.

[0062] The matching unit 315 matches the registered image of the distinguishing part selected by the selection unit 212 with the palm print image while rotating at least one of the registered image and the palm print image (step S33). The matching unit 315 may perform matching each time the inclination changes. [3-3: Technical Effects of Information Processing Device 3]

[0063] For example, if the angle of the object to be compared is not specified, the object to be compared must be rotated within a 360-degree range from -180 degrees to 180 degrees for comparison. The information processing device 3 according to this disclosure not only narrows down the object but also limits the rotation, thereby reducing the processing load. By inferring the angle, the range of the rotation angle can be narrowed, allowing for more efficient comparison. [4: Fourth Embodiment]

[0064] A fourth embodiment of an information processing device, an information processing method, and a recording medium will be described. Hereinafter, the fourth embodiment of an information processing device, an information processing method, and a recording medium will be described using an information processing device 4 according to this disclosure. [4-1: Configuration of Information Processing Device 4]

[0065] The information processing device 4 is configured as a device for learning the estimation operation of the estimation unit 411. More specifically, the information processing device 4 is configured as a device for learning the inference operation of the computation model M.

[0066] The information processing device 4 is configured as a device for selecting distinguishing parts, similar to the information processing devices 1 to 3. The information processing device 4 is also configured as a device for estimating the orientation of a palm print image, similar to the information processing device 3. The information processing device 4 may also be configured as a device for matching palm prints, similar to the information processing devices 2 and 3.

[0067] As shown in Fig. 10, in addition to an estimation unit 411, a selection unit 412, and an acquisition unit 413, a learning unit 417 is further implemented within the calculation device 11 in the fourth embodiment. The estimation unit 411 in the fourth embodiment includes a probability estimation unit 4111 and an inclination estimation unit 4112, similar to the estimation unit 311 in the third embodiment. In addition, a matching unit 315 and a rotation control unit 316 may also be implemented within the calculation device 11 in the fourth embodiment. [4-2: Information Processing Method Executed by Information Processing Device 4]

[0068] The learning unit 417 performs at least one of learning regarding probability estimation by the probability estimation unit 4111 and learning regarding gradient estimation by the gradient estimation unit 4112. In other words, the learning unit 417 performs at least one of learning probability inference operations by the computation model M and learning gradient inference operations. The learning by the learning unit 417 may include learning of parameters of a neural network included in the computation model M.

[0069] The information processing operation in the information processing device 4 will be described with reference to Fig. 11. Fig. 11 is a flowchart showing the flow of the information processing operation in the information processing device 4.

[0070] As shown in Fig. 11, the acquisition unit 413 acquires training data (step S41). The computational model M is constructed by machine learning using the training data. The training data includes a training palm print image (referred to as a "training palm print image") and correct answer information having at least one of part information indicating one or more distinguishing parts included in the training palm print image and tilt information indicating the tilt of the training palm print image. The training palm print image may also be referred to as a multi-labeled image.

[0071] The computation model M is trained so that when a training palmprint image is input, it can output information relating to the correct answer information. When a training palmprint image is input, the computation model M is trained so as to highly infer the probability of one or more distinct parts contained in the training palmprint image. In other words, the computation model M is trained so as to output a probability equal to or greater than a predetermined value for a distinct part contained in the palmprint image. Furthermore, when a training palmprint image is input, the computation model M is trained so as to infer the gradient indicated by the correct answer information.

[0072] The probability estimation unit 4111 estimates the probability that the palm print image contains a distinctive portion using the calculation model M (step S42). The tilt estimation unit 4112 estimates the probability that the distinctive portion is included for each of the multiple distinctive portions.

[0073] The selection unit 412 selects a distinctive part included in the palm print image (step S43). The selection unit 412 may select a distinctive part corresponding to a probability equal to or greater than a predetermined value as a distinctive part included in the palm print image.

[0074] The selection unit 412 may be configured to have a preset upper limit on the number of distinctive portions that can be selected. When the distinctive portions are defined as shown in Fig. 3, the palm print image may include a maximum of three distinctive portions. In this case, the selection unit 412 may be configured to allow selection of three or fewer distinctive portions.

[0075] The tilt estimation unit 4112 uses the calculation model M to estimate the tilt of the palm print image relative to the forward direction of the palm print (step S44).

[0076] The learning unit 417 constructs a computation model M based on a comparison between the estimation result of the estimation unit 411 and the correct answer information (step S45). The learning unit 417 compares the distinguishing portion selected by the selection unit 412 with one or more distinguishing portions included in the learning palm-print image indicated by the correct answer information, and compares the tilt estimated by the tilt estimation unit 4112 with the tilt of the learning palm-print image indicated by the correct answer information. The learning unit 417 may learn the parameters of the neural network included in the computation model M so that the computation model M can infer the correct answer indicated by the correct answer information. [4-3: Technical Effects of Information Processing Device 4]

[0077] Since each part of a palm print is adjacent to another, there is a high possibility that multiple parts are included in a single palm print image. The information processing device 4 according to this disclosure builds a mechanism that can estimate multiple relevant parts by assigning multiple labels and having the device learn. In other words, when a palm print image including multiple distinct parts is input, the information processing device 4 can propose multiple parts to be used as matching targets. Therefore, the information processing device 4 can estimate multiple parts included in a palm print image, and can prevent excess or deficiency of matching targets. [5: Fifth Embodiment]

[0078] A fifth embodiment of an information processing device, an information processing method, and a recording medium will be described below. Hereinafter, the fifth embodiment of an information processing device, an information processing method, and a recording medium will be described using an information processing device 5 according to this disclosure. [5-1: Configuration of Information Processing Device 5]

[0079] The information processing device 5 is configured as a device for matching palm prints, similar to the information processing device 2 and the information processing device 3. Furthermore, the information processing device 5 is configured as a device for selecting distinguishing parts, similar to the information processing device 1 to the information processing device 3. Furthermore, the information processing device 5 may be configured as a device for estimating the orientation of a palm print image, similar to the information processing device 3.

[0080] As shown in Fig. 12, the calculation device 11 in the fifth embodiment further includes an allocation unit 518 in addition to the estimation unit 311, the selection unit 212, the acquisition unit 213, and the matching unit 515. The estimation unit 411 in the fifth embodiment may include a probability estimation unit 3111 and an inclination estimation unit 3112, similar to the estimation unit 311 in the third embodiment. The calculation device 11 in the fifth embodiment may also include a rotation control unit 316 and a learning unit 417. [5-2: Information Processing Method Executed by Information Processing Device 5]

[0081] The assigning unit 518 assigns a level of importance to the matching of the distinctive part with the registered image according to the probability that the distinctive part corresponds to the distinctive part. The matching unit 515 performs matching according to the level of importance. The assigning unit 518 may assign a higher level of importance to the higher the probability.

[0082] The matching unit 515 may prioritize matching with registered images of distinct parts depending on the importance of the distinct parts. The matching unit 515 may prioritize matching with registered images of distinct parts with relatively high importance. For example, the matching unit 515 may perform matching with registered images of distinct parts with the highest importance. Furthermore, when the difference between the highest importance and the second highest importance exceeds a predetermined value, the matching unit 515 may perform matching with registered images of distinct parts with the highest importance. [Modification]

[0083] Alternatively, the matching unit 515 may weight each of the matching results corresponding to the selected distinguishing portion according to the importance and integrate the matching results. [5-3: Technical Effects of the Information Processing Device 5]

[0084] The information processing device 5 according to this disclosure assigns importance to the selected distinguishing portions, thereby enabling early determination without impairing matching accuracy. [6: Supplementary Note]

[0085] The above-described embodiments can be further described as, but not limited to, the following supplementary notes. [Supplementary Note 1] An information processing device comprising: probability estimation means for estimating the probability that a palm print image includes a distinct portion for each of a plurality of the distinct portions; and selection means for selecting a distinct portion included in the palm print image from the plurality of distinct portions based on the probability, wherein the distinct portion is a portion of the palm region that is determined by dividing the palm region and that is to be distinguished. [Supplementary Note 2] The information processing device according to Supplementary Note 1, wherein the probability estimation means performs estimation using a computational model, and when the palm print image is input, the computational model outputs, for each of the plurality of distinct portions, a probability that the palm print image includes the distinct portion. [Supplementary Note 3] The information processing device according to Supplementary Note 1, comprising: matching means for matching a registered image of the distinct portion selected by the selection means with the palm print image. [Supplementary Note 4] The information processing device according to Supplementary Note 1, comprising: tilt estimation means for estimating the tilt of the palmprint image with respect to a predetermined direction; and rotation control means for rotating at least one of the palmprint image and the registered image based on the tilt. [Supplementary Note 5] The information processing device according to Supplementary Note 2, comprising: learning means for performing at least one of learning regarding the probability estimation by the probability estimation means and learning regarding the tilt estimation by the tilt estimation means. [Supplementary Note 6] The information processing device according to Supplementary Note 5, wherein the computational model is constructed by machine learning using as training data including a training palmprint image, part information indicating one or more of the distinctive parts included in the training palmprint image, and ground truth information having tilt information indicating the tilt of the training palmprint image. [Supplementary Note 7] The information processing device according to claim 6, wherein the selection means selects the distinctive parts corresponding to the probability equal to or greater than a predetermined value as the distinctive parts included in the palmprint image. [Supplementary Note 8] The information processing device according to Supplementary Note 6, wherein the selection means selects the distinctive parts equal to or less than an upper limit number of selections. [Supplementary Note 9] The information processing device according to Supplementary Note 3, further comprising: an allocation unit that allocates information indicating a level of importance of matching to a matching of the distinctive part with the registered image in accordance with the probability of the distinctive part; and the matching unit performs matching in accordance with the level of importance.[Supplementary Note 10] The information processing device according to Supplementary Note 9, wherein the matching means prioritizes matching of the distinct part having a relatively high level of importance with the registered image. [Supplementary Note 11] The information processing device according to Supplementary Note 9, wherein the matching means matches the distinct part having the highest level of importance with the registered image. [Supplementary Note 12] The information processing device according to Supplementary Note 9, wherein, when the difference between the highest level of importance and the second highest level of importance exceeds a predetermined value, the matching means matches the distinct part having the highest level of importance with the registered image. [Supplementary Note 13] The information processing device according to Supplementary Note 9, wherein the matching means weights each of the matching results corresponding to the selected distinct parts according to the importance and integrates each of the matching results. [Supplementary Note 14] The information processing device according to Supplementary Note 1, wherein the plurality of distinct parts include the base of the fingers, the ball of the thumb, the ball of the hypothenar, and the palm of the hand, and the base of the fingers, the ball of the thumb, the ball of the hypothenar, and the palm of the hand. [Supplementary Note 15] The information processing device according to Supplementary Note 1, wherein the distinguished parts are determined by dividing the palm area by a first line connecting the metacarpophalangeal joints and a second line extending from the first line and perpendicular to the first line. [Supplementary Note 16] The information processing device according to Supplementary Note 4, wherein the predetermined direction is the direction of a line perpendicular to the first line. [Supplementary Note 17] The information processing device according to Supplementary Note 4, wherein the rotation control means limits the range of rotation based on the tilt. [Supplementary Note 18] An information processing method executed by a computer, which estimates the probability that a palm print image includes a distinguished part for each of a plurality of the distinguished parts, and selects the distinguished part included in the palm print image from the plurality of the distinguished parts based on the probability, wherein the distinguished part is a part in the palm area to be distinguished, determined by dividing the palm area. [Supplementary Note 19] An information processing method for estimating the probability that a palm print image includes a distinct portion for each of a plurality of the distinct portions, and selecting the distinct portion included in the palm print image from among the plurality of the distinct portions based on the probability, wherein the distinct portion is a portion of the palm region that is determined by dividing the palm region and that is to be distinguished. A recording medium having recorded thereon a computer program for causing a computer to execute the information processing method.

[0086] This disclosure may be modified as appropriate within the scope that does not contradict the gist or idea of ​​the invention that can be read from the claims and the entire specification, and information processing devices, information processing methods, and recording media that involve such modifications are also included in the technical idea of ​​this disclosure.

[0087] 1, 2, 3, 4, 5 Information processing device 111, 211, 311, 411 Estimation unit M Computation model 112, 212, 312, 412 Selection unit 213, 413 Acquisition unit 214, 315, 515 Collation unit 3111, 4111 Probability estimation unit 3112, 4112 Inclination estimation unit 316 Rotation control unit 417 Learning unit 518 Allocation unit

Claims

1. An information processing device comprising: a probability estimation means for estimating the probability that a palm print image includes a distinct portion for each of a plurality of distinct portions; and a selection means for selecting a distinct portion included in the palm print image from the plurality of distinct portions based on the probability, wherein the distinct portion is a portion of the palm region that is to be distinguished and is determined by dividing the palm region.

2. The information processing device according to claim 1, wherein the probability estimation means performs estimation using a computational model, which, when the palm print image is input, outputs, for each of the plurality of distinctive features, the probability that the palm print image includes the distinctive feature.

3. The information processing device according to claim 1, further comprising: a matching means for matching the registered image of the distinctive part selected by the selection means with the palm print image.

4. An information processing device according to claim 3, comprising: an inclination estimation means for estimating the inclination of the palm print image relative to a predetermined direction; and a rotation control means for rotating at least one of the palm print image and the registered image based on the inclination.

5. The information processing device according to claim 4, further comprising: learning means for performing at least one of learning regarding the estimation of the probability by the probability estimation means and learning regarding the estimation of the inclination by the inclination estimation means.

6. The information processing device according to claim 5, wherein the computational model is constructed by machine learning using as training data a training palmprint image, part information indicating one or more of the distinguishing parts contained in the training palmprint image, and correct answer information having inclination information indicating the inclination of the training palmprint image.

7. The information processing device according to claim 6, wherein the selection means selects the distinctive portion corresponding to the probability equal to or greater than a predetermined value as the distinctive portion included in the palm print image.

8. The information processing device according to claim 6, wherein the selection means selects the number of distinguishing portions equal to or less than an upper limit of the number of selections.

9. An information processing device according to claim 3, further comprising an allocation means for allocating information indicating the importance of matching to the matching of the distinctive part with the registered image according to the probability of the distinctive part, and wherein the matching means performs matching according to the importance.

10. The information processing device according to claim 9, wherein the matching means prioritizes matching of the distinctive parts having a relatively high degree of importance with the registered images.

11. The information processing device according to claim 9, wherein the matching means matches the registered image of the distinguishing part having the highest importance.

12. The information processing device according to claim 9, wherein when the difference between the highest importance and the second highest importance exceeds a predetermined value, the matching means matches the registered image of the distinguishing part having the highest importance.

13. The information processing device according to claim 9, wherein said matching means weights each of the matching results corresponding to the selected distinguishing parts in accordance with the importance and integrates each of the matching results.

14. The information processing device according to claim 1, wherein the plurality of distinguishable regions include the base of the fingers, the ball of the thumb, the ball of the little finger, and the palm of the hand on the left side, and the base of the fingers, the ball of the thumb, the ball of the little finger, and the palm of the hand on the right side.

15. The information processing device according to claim 1, wherein the distinguishing portion is determined by dividing the palm area by a first line connecting the metacarpophalangeal joints and a second line extending from the first line and perpendicular to the first line.

16. The information processing device according to claim 15, wherein the predetermined direction is the direction of a line perpendicular to the first line.

17. The information processing device according to claim 4, wherein the rotation control means limits the range of rotation based on the tilt.

18. An information processing method executed by a computer, which estimates the probability that a palm print image includes a distinct portion for each of a plurality of distinct portions, and selects, based on the probability, a distinct portion included in the palm print image from among the plurality of distinct portions, wherein the distinct portion is a portion of the palm region that is to be distinguished and is determined by dividing the palm region.

19. A recording medium having recorded thereon a computer program for causing a computer to execute an information processing method comprising: estimating the probability that a palm print image contains a distinct portion for each of a plurality of distinct portions; and selecting, based on the probability, a distinct portion included in the palm print image from among the plurality of distinct portions, wherein the distinct portion is a portion of the palm region that is determined by dividing the palm region and that is to be distinguished.

Images