Information processing device, information processing method, and recording medium

Abstract

An information processing device 1 comprises: an acquisition unit 11 that acquires fingerprint information indicating a plurality of fingerprints corresponding respectively to a plurality of fingers of a subject; a generation unit 12 that classifies each of the plurality of fingerprints according to whether each of the plurality of fingerprints corresponds to a first class, and generates classification information; a collation unit 13 that collates registered registration classification information of the subject with the classification information; and an output unit 14 that outputs a collation result obtained by the collation unit 13.

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] Patent Document 1 describes a technology in which fingerprints are classified into four types: arch (A), left loop (L), right loop (R), and whorl (W), and the feature vector of the fingerprint to be recognized is input to an A recognition unit, an L recognition unit, an R recognition unit, and a W recognition unit, recognition processing is performed by an A matching unit, an L matching unit, an R matching unit, and a W matching unit, and the four types of recognition results are integrated and output.

[0003] Japanese Patent Application Publication No. 09-179847

[0004] An object of this disclosure is to provide an information processing device, an information processing method, and a recording medium that aim to improve upon the techniques disclosed in prior art documents.

[0005] One aspect of the information processing device includes an acquisition means for acquiring fingerprint information indicating a plurality of fingerprints corresponding to each of a plurality of fingers of a subject, a generation means for classifying each of the plurality of fingerprints according to whether it corresponds to a first class or not and generating classification information, a comparison means for comparing the classification information with registered classification information already registered by the subject, and an output means for outputting the comparison result by the comparison means.

[0006] One aspect of the information processing method involves acquiring fingerprint information indicating multiple fingerprints corresponding to each of a subject's multiple fingers, classifying each of the multiple fingerprints based on whether it belongs to a first class, generating classification information, comparing the classification information with registered classification information already registered for the subject, and outputting the comparison result.

[0007] One aspect of the recording medium has recorded thereon a computer program for causing a computer to execute an information processing method that acquires fingerprint information indicating multiple fingerprints corresponding to each of a subject's multiple fingers, classifies each of the multiple fingerprints according to whether it falls into a first class or not, generates classification information, compares the classification information with registered classification information already registered by the subject, and outputs the comparison result.

[0008] FIG. 1 is a block diagram showing an example of the configuration of an information processing device according to the present disclosure. FIG. 2 is a block diagram showing an example of the configuration of an information processing device according to the present disclosure. FIG. 3 is a flowchart showing an example of information processing operation of an information processing device according to the present disclosure. FIG. 4 is a table showing a comparative example of information processing operation of an information processing device according to the present disclosure. FIG. 5 is a block diagram showing an example of the configuration of an information processing device according to the present disclosure. FIG. 6 is a block diagram showing an example of the configuration of an information processing device according to the present disclosure. FIG. 7 is a block diagram showing an example of the configuration of an information processing device according to the present disclosure.

[0009] Hereinafter, embodiments of an information processing device, an information processing method, and a recording medium 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] 1 is a block diagram showing the configuration of an information processing device 1 according to this disclosure. As shown in FIG. 1, the information processing device 1 includes an acquisition unit 11, a generation unit 12, a matching unit 13, and an output unit 14.

[0012] The acquisition unit 11 acquires fingerprint information indicating a plurality of fingerprints corresponding to each of a plurality of fingers of a subject. The generation unit 12 classifies each of the plurality of fingerprints according to whether it corresponds to a first class or not, and generates classification information. The matching unit 13 matches the classification information with the registered classification information already registered for the subject. The output unit 14 outputs the matching result by the matching unit 13. [1-2: Technical Effects of the Information Processing Device 1]

[0013] The information processing device 1 according to this disclosure can identify a subject based on fingerprint information acquired from the subject in a state where it is difficult to obtain an accurate fingerprint. [2: Second Embodiment]

[0014] 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: Individual Identification of Newborns]

[0015] In this embodiment, the subject may be a newborn. In this embodiment, a newborn may refer to a very premature baby within a few hours of birth. For example, in this embodiment, a newborn may refer to a baby within half a day of birth.

[0016] Fingerprints taken from newborns are biometric information that can be used to identify individuals. Furthermore, fingerprints are often stable biometric information because they rarely change with growth, etc.

[0017] On the other hand, the surface of a newborn baby's body is often covered with sebum, amniotic fluid, and the like, making it difficult to obtain fingerprints that can accurately identify the newborn. For example, if the birth occurs in a situation where postpartum care is likely to be inadequate or medical facilities are inadequate, it is often difficult to obtain fingerprints that can accurately identify the newborn. Alternatively, there is a school of thought that recommends delaying the start of bathing a newborn until six hours after birth, in consideration of the newborn's health. In this case, too, it is often difficult to obtain fingerprints that can accurately identify the newborn.

[0018] Therefore, it is difficult to apply fingerprint authentication using minutiae to a newborn baby and accurately identify the newborn baby. In fingerprint authentication using minutiae, it is important to ensure the quality of the data used for authentication, so authentication using low-quality data is rejected as an FTA (Failure to Acquire). Since the body surface of a newborn baby shortly after birth is often covered with sebum, amniotic fluid, etc., there is a high possibility that the newborn baby will be rejected as an FTA in fingerprint authentication using minutiae, making it difficult to authenticate the newborn baby. Furthermore, for example, in the case of a premature birth, the newborn baby's fingerprints may not be fully developed.

[0019] The subject may be a newborn up to four weeks old. The subject may also be an infant, a toddler, or an adult. For example, the subject may be an adult from whom precise fingerprints are difficult to obtain. [2-2: Fingerprint Type]

[0020] The fingerprints of each human finger can be classified into five fingerprint types, for example, Whorl (W), Double Loop (D), Left Loop (L), Right Loop (R), and Arch (A). Similarly, the fingerprints of each finger of a newborn baby can be classified into five fingerprint types, Whorl (W), Double Loop (D), Left Loop (L), Right Loop (R), and Arch (A). Note that these five fingerprint types are merely examples, and other classifications may be adopted.

[0021] A whorl (W) is a fingerprint that has a whorl or ring ridge in the center. A whorl (W) often has deltas on both sides of the central whorl or ring ridge.

[0022] A separation fingerprint (D) is a fingerprint that has a double circular ridge in the center. A separation fingerprint (D) often has deltas on both the left and right sides of the central circular ridge.

[0023] A left loop (L) often has a delta on the right side of the loop-shaped ridge. A left loop (L) is a fingerprint with a loop-shaped ridge that flows from the left side toward the center.

[0024] A right loop (R) is a fingerprint with a loop-shaped ridge that flows from the right side toward the center. A right loop (R) often has a delta on the left side of the loop-shaped ridge.

[0025] An arched fingerprint (A) is a fingerprint having an arched ridge. An arched fingerprint (A) often does not have a delta. [2-3: Configuration of Information Processing Device 2]

[0026] 2 is a block diagram showing the configuration of the information processing device 2. As shown in FIG. 2, the information processing device 2 includes a calculation device 21 and a storage device 22. The information processing device 2 may further include a communication device 23, an input device 24, and an output device 25. However, the information processing device 2 does not necessarily have to include at least one of the communication device 23, the input device 24, and the output device 25. The calculation device 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.

[0027] The arithmetic device 21 includes, for example, at least one of a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), and an FPGA (Field Programmable Gate Array). The arithmetic device 21 reads a computer program. For example, the arithmetic device 21 may read a computer program stored in the storage device 22. For example, the arithmetic device 21 may read a computer program stored in a computer-readable, non-transitory recording medium using a recording medium reading device (e.g., an input device 24 described later) not shown in the drawings that is included in the information processing device 2. The arithmetic device 21 may acquire (i.e., download or read) the computer program from a device (not shown) located outside the information processing device 2 via the communication device 23 (or another communication device). The arithmetic device 21 executes the read computer program. As a result, logical functional blocks for executing operations to be performed by the information processing device 2 are realized within the arithmetic device 21. In other words, the arithmetic device 21 can function as a controller for realizing logical functional blocks for executing operations (in other words, processes) to be performed by the information processing device 2. The arithmetic device 21 may output information to devices (not shown), such as other computers or cloud servers, that are provided outside the information processing device 2, via the communication device 23 (or other communication devices).

[0028] 2 shows an example of logical functional blocks implemented within the computing device 21 to perform information processing operations. As shown in FIG. 3 , the computing device 21 implements an acquisition unit 211, which is a specific example of an "acquisition means" described in the appendix below, a generation unit 212, which is a specific example of a "generation means" described in the appendix below, a matching unit 213, which is a specific example of a "matching means" described in the appendix below, and an output unit 214, which is a specific example of an "output means" described in the appendix below. The generation unit 212 may include a classification unit 2121 and a connection unit 2122. Details of the operations of the acquisition unit 211, the generation unit 212, the matching unit 213, and the output unit 214 will be described later with reference to FIG. 3 .

[0029] The storage device 22 can store desired data. For example, the storage device 22 may temporarily store a computer program executed by the arithmetic device 21. The storage device 22 may temporarily store data that the arithmetic device 21 temporarily uses when the arithmetic device 21 is executing a computer program. The storage device 22 may store data that the information processing device 2 stores for a long period of time. The storage device 22 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 22 may include a non-transitory recording medium. A registration classification information table 221 may be implemented within the storage device 22.

[0030] The communication device 23 is capable of communicating with devices external to the information processing device 2 via a communication network (not shown). The communication device 23 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).

[0031] The input device 24 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 24 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 24 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.

[0032] The output device 25 is a device that outputs information to the outside of the information processing device 2. For example, the output device 25 may output information as an image. That is, the output device 25 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 25 may output information as sound. That is, the output device 25 may include an audio device (a so-called speaker) that can output sound. For example, the output device 25 may output information on paper. That is, the output device 25 may include a printing device (a so-called printer) that can print desired information on paper. [2-4: Information Processing Operation Performed by the Information Processing Device 2]

[0033] The information processing operation performed by the information processing device 2 will be described with reference to Fig. 3. Fig. 3 is a flowchart showing an example of the flow of the information processing operation performed by the information processing device 2.

[0034] As shown in FIG. 3 , the acquisition unit 211 acquires fingerprint information indicating multiple fingerprints corresponding to each of the newborn's multiple fingers (step S20). In this embodiment, a case where fingerprint information indicating fingerprints corresponding to each of the newborn's four fingers is acquired will be described. By collecting fingerprints from four fingers, the burden on the newborn can be reduced. Furthermore, by collecting fingerprints from four fingers, the person can be identified with high accuracy. The four fingers may be, for example, the thumb, index finger, thumb, and index finger of the left hand. The pitch of a newborn's fingerprint is much finer than that of an adult. Therefore, the acquisition unit 211 may acquire a high-resolution fingerprint image.

[0035] The classification unit 2121 classifies each of the four fingerprints based on whether or not they belong to the first class (step S21). The classification unit 2121 may assign "1" to fingerprints that belong to the first class and "0" to fingerprints that do not belong to the first class. In other words, the classification unit 2121 may assign "1" or "0" to each fingerprint.

[0036] If a fingerprint type with a high appearance frequency is assigned to the first class, the appearance frequency of fingerprints that fall into the first class will not be extremely different from the appearance frequency of fingerprints that do not fall into the first class. Therefore, it is preferable to assign a fingerprint type with a high appearance frequency to the first class.

[0037] As an example, there is a theory that Japanese fingerprints most often fall into the whorl (W) type, with half of all fingerprints falling into the whorl (W) type. In this case, the fingerprint type belonging to the first class may be, for example, the whorl (W). Also, for example, in addition to the whorl (W), a separation (D) may also be a fingerprint pattern belonging to the first class. This can also be rephrased as a fingerprint type belonging to the first class having deltas on both the left and right sides of a central ridge. Alternatively, one or more fingerprint types may be assigned to the first class so that, for example, more than 30% of fingerprints fall into the first class.

[0038] The classification unit 2121 may classify whether or not an object belongs to the first class using a classification model constructed by machine learning (for example, a classification model using a neural network). The classification unit 2121 may classify whether or not an object belongs to the first class using a classification model constructed by deep learning, which is an example of machine learning.

[0039] When constructing a classification model, if fingerprint types for which samples are easy to collect belong to the first class, it becomes easy to collect data to be used for training the classification model. From this perspective, it is preferable to assign fingerprint types with a high occurrence frequency to the first class. By assigning fingerprint types with a high occurrence frequency to the first class, it is possible to construct a classification model that can accurately classify whether or not a fingerprint belongs to the first class. Furthermore, by reducing the number of fingerprint types that belong to the first class, the classification model can classify whether or not a fingerprint belongs to the first class with a relatively small amount of learning and training. Furthermore, since it is only necessary to prepare samples of a limited number of fingerprint types, annotation costs can be reduced.

[0040] This embodiment is intended to be applied when it is difficult to obtain an accurate fingerprint. For example, in this embodiment, the fingerprint information may be information obtained from an inked fingerprint. In this case, the fingerprint information does not include information on the side of the finger. Therefore, it is preferable to classify fingerprint types, such as whorl patterns (W), that can be determined to correspond to a fingerprint type even without information on the side of the finger, into the first class.

[0041] For example, since an impression fingerprint does not contain information about the side of the finger, an impression fingerprint of a separated print (D) may be determined to be a left loop (L) or a right loop (R). Therefore, using an impression fingerprint to determine whether or not a separated print (D) is present may hinder improvement in authentication accuracy. Furthermore, since the separated print (D) appears infrequently and it is relatively difficult to collect samples, it is relatively difficult to build a mechanism that can accurately classify whether or not a fingerprint corresponds to a separated print (D). Furthermore, it is difficult to learn a classification method for fingerprints that are difficult to determine, such as variant prints. Therefore, it is preferable to assign fingerprint types that are relatively easy to classify to the first class.

[0042] The concatenation unit 2122 generates classification information by arranging each of the classification results corresponding to each of the multiple fingerprints (step S22). For example, if the classification unit 2121 assigns "1" to the fingerprint of the left thumb, "0" to the left index finger, "0" to the right thumb, and "1" to the right index finger, the concatenation unit 2122 may generate classification information of "1001." The concatenation unit 2122 may generate unique data of the subject by concatenating each of the binary values ​​corresponding to each of the specified fingers.

[0043] It should be noted that which fingers are designated is arbitrary. Any finger can be designated depending on the purpose and situation. In this embodiment, an example is given in which four fingers are designated, but the designated fingers can be changed and any finger can be designated depending on the purpose and situation. For example, all fingers (10 fingers) on both hands may be designated. Also, multiple fingers (2 to 5 fingers) on one hand may be designated. In other words, fingerprint information may be acquired from only one hand of the subject.

[0044] The matching unit 213 acquires the registered registered classification information of the newborn for which classification information has been generated from the registered classification information table 221 (step S23). The registered classification information table 221 is a table in which registered classification information is stored. Like the classification information, the registered classification information is information generated by arranging the classification results. The registered classification information may have the same structure as the classification information. For example, if the registered classification information is connected information of two values ​​for each of the four fingers, the registered classification information may be four-digit binary data. The registered classification information table 221 stores the registered classification information and identification information associated with the newborn in association with each other. The matching unit 213 acquires the registered registered classification information from the registered classification information table 221 based on the identification information. The identification information associated with the newborn will be described later in another embodiment.

[0045] The matching unit 213 matches the classification information with the registered classification information already registered for the newborn (step S24). The matching unit 213 may determine whether the sequence of numbers indicated by the registration classification information matches the sequence of numbers indicated by the classification information. The output unit 214 outputs the matching result by the matching unit 213 (step S25). The output unit 214 may output the matching result by the matching unit 213 to a device (not shown), such as another computer or cloud server, provided outside the information processing device 2, via the communication device 23 (or another communication device). The output unit 214 may control the output device 25 to output the matching result by the matching unit 213. [2-5-1: Modification 1]

[0046] The generation unit 212 may convert classification information expressed in binary by arranging the classification results (1 or 0) corresponding to each of a plurality of fingerprints into classification information expressed in decimal. For example, the generation unit 212 may express classification information expressed in binary as "1001" as "9" in decimal. In this case, the matching unit 213 may match the registered classification information expressed in decimal with the classification information expressed in decimal. [2-5-2: Modification 2]

[0047] The classification unit 2121 may first determine which of the five fingerprint types (whorl (W), detached (D), left loop (L), right loop (R), and arch (A)) each fingerprint corresponds to. The classification unit 2121 may determine whether each fingerprint corresponds to one of the five fingerprint types or to an unknown type (Unknown: U) that does not correspond to any of the five fingerprint types. Next, the classification unit 2121 may determine whether the corresponding fingerprint type belongs to a first class.

[0048] In this case, the classification unit 2121 may use a determination model constructed by machine learning (for example, a determination model using a neural network) to determine which of the above five (or any) fingerprint types the fingerprint belongs to. The classification unit 2121 may use a determination model constructed by deep learning, which is an example of machine learning, to determine which of the above five (or any) fingerprint types the fingerprint belongs to. [2-6: Comparison with Comparative Example]

[0049] 4 shows an example of the result of matching the registered information for five newborns with the information generated during matching. In FIG. 4, the five newborns are identified by the identification information "ID1," "ID2," "ID3," "ID4," and "ID5."

[0050] 4A shows the matching results when each fingerprint is classified into the five fingerprint types described above (referred to as a "comparative example"). In the comparative example, the information that combines the classification results is referred to as "type information." FIG. 4B shows the matching results when this embodiment is applied. In other words, it shows the matching results when each fingerprint is classified according to whether it belongs to the first class.

[0051] In both the comparative example and the case where this embodiment is applied, fingerprints are collected from four fingers of each newborn baby and information processing operations are performed. Also, the information shown in the upper row for each newborn baby may be information obtained during registration, and the information shown in the lower row may be information obtained during matching. [2-6-1: Comparison of newborns with identification information "ID1"] [In the case of the comparative example]

[0052] FIG. 4( a) illustrates an example in which, during registration, the right thumb RT is classified as a whorl (W), the right index finger RI is classified as a whorl (W), the left thumb LT is classified as a whorl (W), and the left index finger LI is classified as a whorl (W), and type information represented as "WWWW" is generated. FIG. 4( a) also illustrates an example in which, during matching, the right thumb RT is classified as a separated print (D), the right index finger RI is classified as a whorl (W), the left thumb LT is classified as a separated print (D), and the left index finger LI is classified as a whorl (W), and type information represented as "DWDW" is generated. Therefore, the type information of the newborn baby for the identification information "ID1" does not match, and the matching result is "FALSE." [In this embodiment]

[0053] FIG. 4(b) illustrates that during registration, the right thumb RT is classified into the first class (1), the right index finger RI is classified into the first class (1), the left thumb LT is classified into the first class (1), and the left index finger LI is classified into the first class (1), generating registration classification information represented as "1111." FIG. 4(b) also illustrates that during matching, the right thumb RT is classified into the first class (1), the right index finger RI is classified into the first class (1), the left thumb LT is classified into the first class (1), and the left index finger LI is classified into the first class (1), generating classification information represented as "1111." Therefore, the classification information of the newborn with identification information "ID1" matches, and the matching result is "TRUE." Note that in FIG. 4, cells that contributed to increasing the number of correct answers are shaded. [2-6-2: Comparison of newborns with identification information "ID2"] [Comparative example]

[0054] FIG. 4( a) illustrates that during enrollment, the right thumb RT is classified as unknown (U), the right index finger RI is classified as left loop (L), the left thumb LT is classified as right loop (R), and the left index finger LI is classified as right loop (R), generating type information represented as "ULRR." FIG. 4( a) also illustrates that during matching, the right thumb RT is classified as unknown (U), the right index finger RI is classified as left loop (L), the left thumb LT is classified as right loop (R), and the left index finger LI is classified as right loop (R), generating type information represented as "ULRR." The newborn's type information for the identification information "ID2" matches, but includes unknown (U), so the result is "FTE (Fair to enroll)." [In this embodiment]

[0055] 4(b) illustrates that during enrollment, the thumb RT of the right hand is classified into a class other than the first class (0), the index finger RI of the right hand is classified into a class other than the first class (0), the thumb LT of the left hand is classified into a class other than the first class (0), and the index finger LI of the left hand is classified into a class other than the first class (0), and thus enrollment classification information represented as "0000" is generated. Also, FIG. 4(b) illustrates that during matching, the thumb RT of the right hand is classified into a class other than the first class (0), the index finger RI of the right hand is classified into a class other than the first class (0), the thumb LT of the left hand is classified into a class other than the first class (0), and the index finger LI of the left hand is classified into a class other than the first class (0), and thus classification information represented as "0000" is generated. Therefore, the classification information of the newborn with the identification information "ID2" matches, and the matching result is "TRUE." [2-6-3: Comparison of newborns with identification information "ID3"] [Comparative example]

[0056] FIG. 4( a) illustrates that, during registration, the right thumb RT is classified as a left loop (L), the right index finger RI is classified as a separate loop (D), the left thumb LT is classified as a separate loop (D), and the left index finger LI is classified as a whorl (W), generating type information represented as "LDDW." FIG. 4( a) also illustrates that, during matching, the right thumb RT is classified as a left loop (L), the right index finger RI is classified as a whorl (W), the left thumb LT is classified as a whorl (W), and the left index finger LI is classified as a whorl (W), generating type information represented as "LWWWW." Therefore, the type information of the newborn baby for the identification information "ID3" does not match, and the matching result is "FALSE." [In this embodiment]

[0057] 4(b) illustrates that during enrollment, the thumb RT of the right hand is classified into a class other than the first class (0), the index finger RI of the right hand is classified into the first class (1), the thumb LT of the left hand is classified into the first class (1), and the index finger LI of the left hand is classified into the first class (1), and thus enrollment classification information represented as "0111" is generated. Also, FIG. 4(b) illustrates that during matching, the thumb RT of the right hand is classified into a class other than the first class (0), the index finger RI of the right hand is classified into the first class (1), the thumb LT of the left hand is classified into the first class (1), and the index finger LI of the left hand is classified into the first class (1), and thus classification information represented as "0111" is generated. Therefore, the classification information of the newborn with the identification information "ID3" matches, and the matching result is "TRUE." [2-6-4: Comparison of newborns with identification information "ID4"] [Comparative example]

[0058] FIG. 4( a) illustrates that, during registration, the right thumb RT is classified as a whorl (W), the right index finger RI is classified as a right loop (R), the left thumb LT is classified as a whorl (W), and the left index finger LI is classified as a whorl (W), generating type information represented as "WRWW." FIG. 4( a) also illustrates that, during matching, the right thumb RT is classified as a left loop (L), the right index finger RI is classified as a right loop (R), the left thumb LT is classified as a left loop (L), and the left index finger LI is classified as a left loop (L), generating type information represented as "LRLL." Therefore, the type information of the newborn baby in the identification information "ID4" does not match, and the matching result is "FALSE." [In this embodiment]

[0059] 4(b) illustrates that during enrollment, the thumb RT of the right hand is classified into the first class (1), the index finger RI of the right hand is classified into a class other than the first class (0), the thumb LT of the left hand is classified into the first class (1), and the index finger LI of the left hand is classified into the first class (1), and thus enrollment classification information represented as "1011" is generated. Also, FIG. 4(b) illustrates that during matching, the thumb RT of the right hand is classified into a class other than the first class (0), the index finger RI of the right hand is classified into a class other than the first class (0), the thumb LT of the left hand is classified into a class other than the first class (0), and the index finger LI of the left hand is classified into a class other than the first class (0), and thus classification information represented as "0000" is generated. Therefore, the classification information of the newborn baby with identification information "ID4" does not match, and the matching result is "FALSE." [2-6-5: Comparison of newborns with identification information "ID5"] [Comparative example]

[0060] FIG. 4( a) illustrates that during registration, the right thumb RT is classified as a whorl (W), the right index finger RI is classified as a whorl (W), the left thumb LT is classified as a whorl (W), and the left index finger LI is classified as a whorl (W), and type information represented as "WWWW" is generated. FIG. 4( a) also illustrates that during matching, the right thumb RT is classified as a separated print (D), the right index finger RI is classified as a whorl (W), the left thumb LT is classified as a separated print (D), and the left index finger LI is classified as an arch print (A), and type information represented as "DWDA" is generated. Therefore, the type information of the newborn baby with the identification information "ID5" does not match, and the matching result is "FALSE." [In this embodiment]

[0061] 4(b) illustrates that during registration, the thumb RT of the right hand is classified into the first class (1), the index finger RI of the right hand is classified into the first class (1), the thumb LT of the left hand is classified into the first class (1), and the index finger LI of the left hand is classified into the first class (1), and thus registration classification information represented as "1111" is generated. Also, FIG. 4(b) illustrates that during matching, the thumb RT of the right hand is classified into the first class (1), the index finger RI of the right hand is classified into the first class (1), the thumb LT of the left hand is classified into the first class (1), and the index finger LI of the left hand is classified into a class other than the first class (0), and thus classification information represented as "1110" is generated. Therefore, the classification information of the newborn with the identification information "ID5" does not match, and the matching result is "FALSE."

[0062] In the comparative example, if accurate classification cannot be performed for all five fingerprint types, the result is determined to be "FALSE." Therefore, even in one-to-one authentication, it is difficult to achieve successful authentication.

[0063] In contrast, in this embodiment, whorl prints (W) and separation prints (D) belong to the first class, and fingerprint types other than whorl prints (W) and separation prints (D) are classified as not belonging to the first class. In this embodiment, fingerprints whose fingerprint type is unknown (U) due to reasons such as low quality are also classified as not belonging to the first class, so even if a fingerprint whose fingerprint type is unknown (U) is present, matching can be performed without determining it as "FTE (fail)." Therefore, while the method of the comparative example determined the fingerprint to be "TRUE" 0% of the time, the method of this embodiment can improve the rate of determining the fingerprint to be "TRUE" to 60%.

[0064] In this embodiment, the case where the whorl (W) and the separated (D) belong to the first class has been exemplified, but the fingerprint classifications belonging to the first class may be other than the whorl (W) and separated (D). For example, the left loop (L) may belong to the first class. Also, the right loop (R) may belong to the first class. Also, the left loop (L) and the right loop (R) may belong to the first class. [2-7: Technical Effects of Information Processing Device 2]

[0065] The information processing device 2 according to this disclosure can identify a newborn based on fingerprint information obtained from a newborn, for whom accurate fingerprints are often difficult to obtain. The information processing device 2 performs matching using classification information generated by arranging classification results corresponding to each of a plurality of fingerprints, thereby achieving highly accurate matching with a relatively small amount of information. [3: Third Embodiment]

[0066] 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: Information processing operation performed by the information processing device 3]

[0067] The third embodiment differs from the second embodiment in the operation of the classification unit 3121. The classification unit 3121 sets a fingerprint type belonging to the first class according to an area associated with a newborn as a subject. The area associated with a newborn may be the area where the newborn was born. The area associated with a newborn may be stored in association with identification information associated with the newborn. In this embodiment, the term "area" may refer to an area divided according to an administrative district such as a city or a county. The term "area" may also refer to an area divided according to an environment such as topography, geology, climate, or vegetation. For example, areas within the same administrative district may be divided according to altitude (such as along the coastline or in mountainous areas) and each area may be treated as a separate area. In this embodiment, any area can be distinguished as a "area."

[0068] The area associated with the newborn may be the area where the fingerprint information was acquired. In this case, location information may be linked to the fingerprint information acquired by the acquisition unit 311. The classification unit 3121 may determine the area associated with the newborn based on the location information linked to the fingerprint information.

[0069] The area associated with the newborn may be an area where the information processing device 3 is used. In this case, the classification unit 3121 may determine the area associated with the newborn based on location information of the location where the information processing device 3 is used.

[0070] A region information table 322 may be constructed in the storage device 22 of the information processing device 3. The region information table 322 may store information associating a region with a fingerprint type belonging to a first class. As described above, it is preferable to assign a fingerprint type with a high appearance frequency to the first class. It is also known that the appearance tendency of fingerprint types differs from region to region. The region information table 322 may store information associating a region with a fingerprint type with a high appearance frequency in the region.

[0071] One or more fingerprint types may belong to the first class. That is, multiple fingerprint types may belong to the first class. Therefore, there may be an area associated with one fingerprint type and an area associated with multiple fingerprint types. For example, a whorl (W) may be associated with the first area. In this case, the whorl (W) belongs to the first class in the first area. Furthermore, a left loop (L) and a right loop (R) may be associated with the second area. In this case, the left loop (L) and the right loop (R) belong to the first class in the second area.

[0072] The classification unit 3121 may be able to use a plurality of classification models. Each of the plurality of classification models has a different fingerprint type belonging to the first class. The classification unit 3121 may classify the fingerprint of a newborn baby using a classification model in which a fingerprint type associated with a region related to the newborn baby belongs to the first class. The region information table 322 may store information associating a region with a classification model used by the classification unit 3121 in addition to or instead of information associating a region with a fingerprint type. [3-2: Modification]

[0073] The classification unit 3121 may first determine which of the five fingerprint types (whorl (W), detached (D), left loop (L), right loop (R), and arch (A)) each fingerprint corresponds to. The classification unit 3121 may determine whether each fingerprint corresponds to one of the five fingerprint types or whether it is unknown (U), which is not one of the five fingerprint types. Next, the classification unit 2121 may determine whether the fingerprint type determined to correspond belongs to a first class.

[0074] In this case, the generation unit 312 may use the same classification model for all regions, and further classify the classification results into which of the above five types the image falls into based on whether or not the image falls into the first class. [3-3: Technical Effects of the Information Processing Device 3]

[0075] The information processing device 3 according to this disclosure can improve the accuracy of matching by assigning fingerprint types according to regions to the first class. The information processing device 3 can perform operations suited to the region. [4: Fourth Embodiment]

[0076] A fourth embodiment of an information processing device, an information processing method, and a recording medium will be described below. Hereinafter, the fourth embodiment of the information processing device, the information processing method, and the recording medium will be described using an information processing device 4 according to this disclosure. [4-1: Information processing operation performed by the information processing device 4]

[0077] The fourth embodiment differs from the second and third embodiments in the registered classification information registered in the registered classification information table 421. Furthermore, the fourth embodiment differs from the second and third embodiments in the operations of the generating unit 412 and the matching unit 413.

[0078] The first registration classification information when the first fingerprint type belongs to the first class and the second registration classification information when the second fingerprint type belongs to the first class are already registered in the registration classification information table 421. For example, the first registration classification information when "W" belongs to the first class and the second registration classification information when "R" belongs to the first class may be already registered in the registration classification information table 421.

[0079] The generation unit 412 generates at least one of first classification information when the first fingerprint type belongs to the first class and second classification information when the second fingerprint type belongs to the first class. The generation unit 412 may accept a specification as to whether to perform matching using the first classification information when the first fingerprint type belongs to the first class, whether to perform matching using the second classification information when the second fingerprint type belongs to the first class, or whether to perform matching using both the first classification information and the second classification information. In this case, the generation unit 412 may accept a specification as to which classification information to use for matching via, for example, the input device 24.

[0080] The matching unit 413 performs at least one of a first matching between the first registered classification information and the first classification information and a second matching between the second registered classification information and the second classification information. The matching unit 413 may perform the matching in accordance with the classification information generated by the generation unit 412. [4-2: Technical Effects of the Information Processing Device 4]

[0081] The information processing device 4 according to this disclosure can identify a target person with high accuracy. The information processing device 4 can identify a target person by matching according to the purpose and situation. [5: Fifth Embodiment]

[0082] A fifth embodiment of an information processing device, an information processing method, and a recording medium will be described. 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: Identification Information]

[0083] The identification information may be biometric information of a person related to the subject. If the subject is a newborn, the identification information may be biometric information of the mother who gave birth to the newborn (hereinafter, sometimes simply referred to as "mother"). When associating a newborn with a person related to the newborn, associating the newborn with the mother who gave birth to the newborn is the most reliable and useful for identifying the newborn. For example, a mother who gave birth to twins and triplets born one year apart will have five corresponding children under the age of one. Since it is rare for a single mother to have more than five corresponding children, it can be assumed that one-to-f (f is 5 or less) matching will be possible. Since it is extremely rare for the biometric information of five people to have the same pattern, associating the newborn with the mother is useful for identifying the newborn. Furthermore, according to the information available to the applicant at the time of filing the application, there is a case in which 69 people gave birth, including four sets of quadruplets, the world's highest number of pregnancies and births in human history, and the world record for the highest number of multiple births in human history was 10 triplets, so one-to-f (f is 10 or less) matching may be possible. Since it is extremely rare for the biological information of 10 people to have the same pattern, associating a newborn with its mother is useful for identifying the newborn. Furthermore, in many cases where newborn information is used, it is appropriate to associate the newborn with its mother.

[0084] The biological information of the mother may be, for example, information on the fingerprint of the newborn's mother, information on the mother's voice, information on the mother's face, or information on the mother's iris.

[0085] Alternatively, the identification information may be information contained in an object possessed by the subject. The identification information may be ID data that can be read from a card carried by the subject. In this case, the matching of the classification information is one-to-one matching. [5-2: Information Processing Operation Performed by Information Processing Device 5]

[0086] When the identification information is biometric information of a person associated with the subject, the matching unit 513 first performs biometric authentication on the person associated with the subject, and acquires the registered registration classification information of the subject associated with the person's biometric information from the registration classification information table 521. When the identification information is information contained in an object owned by the subject, the matching unit 513 first acquires the registered registration classification information of the newborn associated with ID data or the like from the registration classification information table 521. [5-3: Technical Effects of Information Processing Device 5]

[0087] The information processing device 5 according to this disclosure can perform highly accurate matching even for newborns, for which it is difficult to collect data that allows for highly accurate matching, by performing identification information matching and pattern matching. In other words, the information processing device 4 can ensure a certain level of matching accuracy by performing one-to-one (or one-to-a relatively small number of) matching using sequence data of newborns associated with the person after performing matching of a person related to the newborn.

[0088] The information processing device 5 registers the registered classification information in association with the identification information related to the subject, so that the newborn can be identified with high accuracy. The information processing device 5 uses the identification information and classification information related to the subject, so that the newborn can be identified with high accuracy even when, for example, biometric information of a certain quality or higher required for biometric authentication of an adult cannot be obtained. [6: Sixth Embodiment]

[0089] A sixth embodiment of an information processing device, an information processing method, and a recording medium will be described. Hereinafter, the sixth embodiment of an information processing device, an information processing method, and a recording medium will be described using an information processing device 6 according to this disclosure. [6-1: Information processing operation performed by the information processing device 6]

[0090] In the sixth embodiment, the operation of the generation unit 612 differs from the second to fifth embodiments. A classification unit 6121 classifies each of a plurality of fingerprints based on whether it belongs to a first class or a second class. The second class is different from the first class.

[0091] The first class may be a class to which fingerprint types having two deltas belong. The first class may be a class to which fingerprint types having deltas on both the left and right sides of a central ridge belong. When fingerprints are classified into the above five fingerprint types, the first class may be a class to which whorls (W) and detached prints (D) belong.

[0092] The second class may be a class to which fingerprint types having one delta belong. The second class may be a class to which fingerprint types having a delta on the left or right side of the central ridge belong. When fingerprints are classified into the above five fingerprint types, the second class may be a class to which left loop (L) and right loop (R) belong.

[0093] In this case, fingerprint types that do not have a delta belong to neither class 1 nor class 2. When classifying fingerprints into the above five fingerprint types, arch (A) and unknown (U) belong to neither class 1 nor class 2.

[0094] The concatenation unit 6122 may arrange the ternary information to generate ternary classification information. Alternatively, the concatenation unit 6122 may express the classification information as decimal numbers. [6-2: Modification]

[0095] The classification unit 6121 first classifies each of the plurality of fingerprints based on whether it belongs to a first class or not, and whether it belongs to a second class or not, and assigns one of three values ​​to each of the plurality of fingerprints. Subsequently, the classification unit 6121 may perform two-class classification based on whether it belongs to a target class or not. [6-3: Technical Effects of the Information Processing Device 6]

[0096] The information processing device 6 according to this disclosure performs matching using classification information generated from the classification results obtained by classifying each fingerprint into three classes, and therefore can improve matching accuracy compared to matching using classification information generated from the classification results obtained by classifying each fingerprint into two classes. [7: Note]

[0097] The following supplementary notes are further disclosed with respect to the above-described embodiments. [Supplementary Note 1] An information processing device comprising: acquisition means for acquiring fingerprint information indicating a plurality of fingerprints corresponding to each of a plurality of fingers of a subject; generation means for classifying each of the plurality of fingerprints according to whether it belongs to a first class and generating classification information; comparison means for comparing the classification information with registered classification information already registered for the subject; and output means for outputting a comparison result by the comparison means. [Supplementary Note 2] The information processing device of claim 1, wherein the subject is a newborn. [Supplementary Note 3] The information processing device of claim 1, wherein the generation means generates the classification information by arranging each of the classification results corresponding to each of the plurality of fingerprints. [Supplementary Note 4] The information processing device of claim 1, wherein the generation means sets the fingerprint type belonging to the first class according to a region associated with the subject. [Supplementary Note 5] The information processing device of claim 1, wherein first registered classification information when a first fingerprint type belongs to the first class and second registered classification information when a second fingerprint type belongs to the first class have already been registered, the generation means generates at least one of the first classification information when the first fingerprint type belongs to the first class and the second classification information when the second fingerprint type belongs to the first class, and the matching means performs at least one of a first matching between the first registered classification information and the first classification information and a second matching between the second registered classification information and the second classification information. [Supplementary Note 6] The information processing device of claim 1, wherein the registered classification information is registered in association with identification information related to the subject. [Supplementary Note 7] The information processing device of claim 1, wherein the generation means classifies each of the plurality of fingerprints based on whether it belongs to the first class and whether it belongs to a second class different from the first class, to generate the classification information. [Supplementary Note 8] An information processing method comprising: acquiring fingerprint information indicating a plurality of fingerprints corresponding to each of a plurality of fingers of a subject; classifying each of the plurality of fingerprints according to whether it corresponds to a first class or not, thereby generating classification information; comparing the classification information with registered classification information already registered for the subject; and outputting a comparison result.[Supplementary Note 9] A recording medium having recorded thereon a computer program for causing a computer to execute an information processing method, which includes: acquiring fingerprint information indicating a plurality of fingerprints corresponding to each of a plurality of fingers of a subject; classifying each of the plurality of fingerprints according to whether it belongs to a first class or not, thereby generating classification information; comparing the classification information with registered classification information of the subject; and outputting a comparison result.

[0098] Although this disclosure has been described above with reference to the embodiments, this disclosure is not limited to the above-described embodiments. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of this disclosure within the scope of this disclosure. Furthermore, each embodiment can be combined with other embodiments as appropriate.

[0099] 1, 2, 3, 4, 5, 6 Information processing device 11, 211 Acquisition unit 12, 212, 312, 412, 612 Generation unit 13, 213, 413, 513, 613 Collation unit 14, 214 Output unit 221, 421, 521 Registered classification information table 2121, 3121, 6121 Classification unit 2122, 6122 Linkage unit 322 Regional information table

Claims

1. An information processing device comprising: an acquisition means for acquiring fingerprint information indicating a plurality of fingerprints corresponding to each of a plurality of fingers of a subject; a generation means for classifying each of the plurality of fingerprints according to whether or not it corresponds to a first class, and generating classification information; a comparison means for comparing the classification information with registered classification information of the subject; and an output means for outputting a comparison result by the comparison means.

2. The information processing device according to claim 1, wherein the subject is a newborn.

3. The information processing device according to claim 1, wherein said generating means generates said classification information by arranging each of the classification results corresponding to each of said plurality of fingerprints.

4. The information processing device according to claim 1, wherein said generating means sets the fingerprint type belonging to said first class according to an area associated with said subject.

5. The information processing device according to claim 1, wherein first registered classification information when a first fingerprint type belongs to the first class and second registered classification information when a second fingerprint type belongs to the first class have already been registered, the generation means generates at least one of the first classification information when the first fingerprint type belongs to the first class and the second classification information when the second fingerprint type belongs to the first class, and the matching means performs at least one of a first matching between the first registered classification information and the first classification information and a second matching between the second registered classification information and the second classification information.

6. The information processing device according to claim 1, wherein the registered classification information is registered in association with identification information related to the subject.

7. The information processing device according to claim 1, wherein the generation means classifies each of the plurality of fingerprints according to whether it corresponds to the first class or not and whether it corresponds to a second class different from the first class or not, thereby generating the classification information.

8. An information processing method comprising: acquiring fingerprint information indicating a plurality of fingerprints corresponding to each of a plurality of fingers of a subject; classifying each of the plurality of fingerprints according to whether it corresponds to a first class or not, generating classification information; comparing the classification information with registered classification information of the subject; and outputting a comparison result.

9. A recording medium having recorded thereon a computer program for causing a computer to execute an information processing method that includes acquiring fingerprint information indicating a plurality of fingerprints corresponding to each of a plurality of fingers of a subject, classifying each of the plurality of fingerprints according to whether or not it corresponds to a first class, generating classification information, comparing the classification information with registered classification information of the subject, and outputting the comparison result.

Images