Information processing device, information processing method, and program

Abstract

To contribute to the provision of an information processing device, an information processing method and a program capable of reducing time required for face authentication.SOLUTION: An information processing device includes a feature amount conversion part for converting a feature amount acquired from a face image and included in a first numerical range into a conversion value in a second numerical range narrower than the first numerical range, and a calculation part for calculating an evaluation value showing similarity to a second conversion value associated with a second feature amount of an object person of face authentication in each first conversion value stored associated with a first feature amount of each registrant registered in advance, and selecting the first feature amount associated with the first conversion value at which the evaluation value is equal to or larger than a threshold.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present disclosure relates to an information processing device, an information processing method, and a program. [Background technology]

[0002] In recent years, the use of face recognition to authenticate people and manage their entrance and exit has been considered. For example, in facilities such as offices and schools where entry is permitted for people who are registered in advance and entry is restricted for people who are not registered, the use of face recognition to authenticate that a person attempting to enter (for example, a person to be authenticated) is a person permitted to enter has been considered.

[0003] For example, Patent Document 1 describes that a face detection process is performed on a captured image, a face image is extracted from a facial area that is thought to be a person's face, the face image is normalized to a predetermined number of pixels, and the similarity between the features in the normalized target face image and the features in a registered face image is calculated as a score, thereby calculating how "like" the target face image is of a specific person. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2013-134736 A Summary of the Invention [Problem to be solved by the invention]

[0005] However, the method described in Patent Document 1 requires calculating a score for each registered face image, which increases the time required for face authentication.

[0006] Non-limiting examples of the present disclosure contribute to providing an information processing device, an information processing method, and a program that can reduce the time required for face authentication. [Means for solving the problem]

[0007] An information processing device according to one embodiment of the present disclosure includes a feature conversion unit that converts a feature acquired from a face image and falling within a first numerical range into a conversion value within a second numerical range narrower than the first numerical range, and a calculation unit that calculates an evaluation value indicating a similarity between a second conversion value associated with a second feature of a subject of facial authentication for each first conversion value that is associated with a first feature of each pre-registered person and stored, and selects a first feature associated with a first conversion value for which the evaluation value is equal to or greater than a threshold value.

[0008] In an information processing method according to one embodiment of the present disclosure, an information processing device converts a feature value obtained from a facial image and falling within a first numerical range into a conversion value within a second numerical range narrower than the first numerical range, calculates an evaluation value indicating the similarity between each first conversion value stored in association with the first feature value of each pre-registered person and a second conversion value associated with the second feature value of a subject of facial authentication, and selects a first feature value associated with a first conversion value for which the evaluation value is equal to or greater than a threshold value.

[0009] A program according to an embodiment of the present disclosure causes an information processing device to execute a process of converting a feature acquired from a facial image and falling within a first numerical range into a conversion value within a second numerical range narrower than the first numerical range, calculating an evaluation value indicating a similarity between each first conversion value stored in association with a first feature of each pre-registered person and a second conversion value associated with a second feature of a subject of facial authentication, and selecting a first feature associated with a first conversion value for which the evaluation value is equal to or greater than a threshold value.

[0010] In addition, these comprehensive or specific aspects may be realized by a system, an apparatus, a method, an integrated circuit, a computer program, or a recording medium, or may be realized by any combination of a system, an apparatus, a method, an integrated circuit, a computer program, and a recording medium. Effect of the Invention

[0011] Non-limiting examples of the present disclosure can reduce the time required for face recognition.

[0012] Further advantages and benefits of an embodiment of the present disclosure will become apparent from the specification and drawings, in which such advantages and / or benefits are provided by some of the embodiments and features described in the specification and drawings, respectively, but not necessarily all of which are provided to obtain one or more identical features. [Brief description of the drawings]

[0013] [Figure 1] A graph showing an example of the relationship between the number of SBMs and the score [Diagram 2] FIG. 1 is a diagram showing an example of the configuration of an information processing device according to an embodiment; [Diagram 3] A diagram showing the first example of registered person features and the SB column [Figure 4] A second example of registered user features and the SB column [Diagram 5] Figure showing example 1 of SBM threshold settings [Figure 6] A diagram showing an example of the relationship between the number of SBMs and the number of cases [Figure 7] A diagram showing an example of the distribution of numbers based on the relationships shown in FIG. [Figure 8] Example 2 of SBM threshold setting [Figure 9] Supplementary figure for the process of SBM threshold setting example 2 [Figure 10] A flowchart showing an example of the flow of registration processing in an information processing device according to an embodiment. [Figure 11] A flowchart showing an example of the flow of a threshold setting process in an information processing device according to an embodiment. [Figure 12] A flowchart showing an example of the flow of a score calculation process in an information processing device according to an embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0014] Hereinafter, preferred embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. In this specification and the drawings, components having substantially the same functions are denoted by the same reference numerals, and redundant description will be omitted.

[0015] (One embodiment) <Knowledge that led to this disclosure> Facial recognition is one of the methods for verifying the identity of a subject of facial recognition (hereinafter, subject). Facial recognition is used to determine whether the subject is a person who has been registered in advance (hereinafter, registrant), and, if multiple registrants are registered, to determine which of the registrants the subject is. In the process of facial recognition, a matching score is calculated that indicates the similarity between a facial feature amount extracted from a facial image of the subject and a facial feature amount extracted from a facial image of a registrant, and a determination is made as to which of the registrants the subject is based on the level of the matching score. The facial feature amount is designed so that the matching score is relatively high when the subject and the registrant are the same person, and the matching score is low when the subject and the registrant are not the same person. Hereinafter, the facial feature amount may be abbreviated to feature amount, and the matching score may be abbreviated to score.

[0016] Since the calculation of the score between the feature amount of the subject and the feature amount of the enrolled person is performed for each enrolled person, the more enrolled people there are, i.e., the more feature amounts of the enrolled people there are, the longer the score calculation process takes.

[0017] Since facial recognition is used to determine who the subject is among the registered users, the score calculation process can be sped up by limiting the targets for score calculation to the features of registered users that are expected to produce high scores.

[0018] Therefore, in this embodiment, a method is described in which the feature amount is converted into another value based on the feature amount, and the converted value is used to determine which feature amount of a registered person is expected to have a high score, thereby limiting the targets for score calculation to the feature amounts of registered people expected to have a high score, thereby speeding up the score calculation process.

[0019] <Example of feature conversion> In the present embodiment, the feature amount is converted into a binary value, for example. Hereinafter, the conversion into a binary value will be referred to as binarization.

[0020] The score is calculated by using an inner product calculation. For example, the inner product calculation of a feature quantity represented by an n-dimensional vector (n is an integer equal to or greater than 1) will be described.

[0021] If two features are written as n-dimensional vectors v0 and v1, and the element of dimension i (i is an integer between 1 and n) of vector v0 is written as v0[i], and the element of dimension i of vector v1 is written as v1[i], then the score is calculated as shown in equation (1).

number

[0022] Here, if v0[i] and v1[i] have the same sign, for example, if the values ​​of v0[i] and v1[i] are both positive or negative, the product of v0[i] and v1[i] will be a positive value. In this case, the score represented by the sum of the products for each element will increase.

[0023] On the other hand, if v0[i] and v1[i] have different signs, for example, if one of v0[i] and v1[i] is positive and the other is negative, the product of v0[i] and v1[i] is negative, and the score represented by the sum of the products for each element decreases.

[0024] When v0[i] and v1[i] have the same sign as v1[i], they are written as v0p[i] and v1p[i]. When v0[i] and v1[i] have a different sign from v1[i], they are written as v0m[i] and v1m[i]. In this case, the score shown in formula (1) is calculated as shown in formula (2).

number

[0025] Here, p indicates the number of pairs of v0p[i] and v1p[i] having the same positive or negative sign, and m indicates the number of pairs of v0m[i] and v1m[i] having different positive or negative signs.

[0026] The fact that the score increases when v0[i] and v1[i] have the same positive or negative sign means that the score is expected to increase as the number of p increases. Hereinafter, if the number of p, that is, the number of pairs of v0p[i] and v1p[i] with the same positive or negative sign, is referred to as the SBM (Sign Bit Match) number, then the greater the SBM number between two features, the greater the score calculated between the two features is expected to be.

[0027] FIG. 1 is a diagram showing an example of the relationship between the number of SBMs and the score. FIG. 1 shows the results of calculating the score between feature amounts extracted from the evaluation images and the number of SBMs between the feature amounts for each pair of 1 million evaluation images. The horizontal axis of FIG. 1 shows the number of SBMs, and the vertical axis shows the score. FIG. 1 shows a graph showing the maximum score for the number of SBMs and a graph showing the minimum score for the number of SBMs. Information showing the maximum score for the number of SBMs is written as maximum score information, information showing the minimum score for the number of SBMs is written as minimum score information, and the maximum score information and the minimum score information may be written together as score information. The score information may be an example of a statistical relationship between the number of SBMs and the score.

[0028] As shown in Figure 1, the greater the number of SBMs between two features, the greater the score calculated between the two features.

[0029] The SBM number between two features is calculated using the binarized features obtained by binarizing the two features according to their positive and negative signs. A vector (a sequence of numbers) consisting of values ​​binarized according to the positive and negative signs is sometimes referred to as an SB sequence (Sign Bit sequence).

[0030] In this embodiment, the feature is binarized, the binarized value is used to calculate the SBM number, and a determination is made to select the feature of a registered user who is expected to have a high score based on the SBM number. Note that, hereinafter, the binarized feature may be referred to as a binarized feature or an SB string. The SBM number may be referred to as the degree of agreement between two binarized features. The degree of agreement between two binarized features corresponds to an example of an evaluation value indicating the similarity between the two binarized features.

[0031] <Device configuration> 2 is a diagram showing an example of the configuration of the information processing device 1 according to the present embodiment. The information processing device 1 shown in FIG.

[0032] The communication unit 11 is an interface that performs a transmission process related to transmitting information to an external device and a reception process related to receiving information from an external device. Here, the external device may be, for example, a device that requests the information device 1 to register information of a registrant in face authentication, and / or a device that requests face authentication from the information device 1. The communication unit 11 may be divided into a transmission unit that performs a transmission process and a reception unit that performs a reception process. Note that the transmission and reception of information correspond to the transmission and reception of a signal including information, respectively. Also, the reception of information corresponds to an example of acquiring information.

[0033] When the information processing device 1 registers information about a registrant, the communication unit 11 receives personal information about the registrant (identification information, name, face image, etc. of the registrant). When the information processing device 1 executes face authentication, the communication unit 11 receives feature amounts of a target person for face authentication.

[0034] The processor 12 operates an operating system to control the entire information processing device 1. The processor 12 may be configured by a central processing unit (CPU) including an interface with peripheral devices, a control device, an arithmetic unit, a register, etc. The processor 12 reads out programs (program codes), software modules, data, etc. from at least one of the storage unit 13 and the communication unit 11 into a memory (omitted in FIG. 2), and executes various processes according to the programs. As the program, a program that causes the information processing device to execute at least a part of the operations described in this embodiment is used. The program may be transmitted from a network via a telecommunication line.

[0035] The processor 12 processes information related to face authentication. The processor 12 includes a feature conversion unit 121, a calculation unit 122, a threshold setting unit 123, and an encryption processing unit .

[0036] The feature conversion unit 121 acquires a feature and binarizes the feature to generate a binarized feature. The relationship between the feature and the binarized feature is that the feature is included in a certain numerical range (for example, a range of real numbers with upper and lower limits defined), whereas the binarized feature is included in a numerical range having only two values ​​narrower than the numerical range of the feature. When the range in which the feature is included is referred to as a first numerical range, the feature conversion unit 121 converts the feature included in the first numerical range into a converted value (for example, a binarized feature) within a numerical range (for example, two values) narrower than the first numerical range.

[0037] For example, when the information processing device 1 registers information of a registrant, the feature conversion unit 121 binarizes the feature of the registrant to generate binary feature of the registrant, and links the feature of the registrant, the binary feature of the registrant, and personal information of the registrant to each other and stores them in the storage unit 13. Also, for example, when the information processing device 1 performs face authentication, the feature conversion unit 121 binarizes the feature of a subject of face authentication to generate binary feature of the subject.

[0038] The calculation unit 122 selects the binary feature of the enrolled person stored in the storage unit 13. The calculation unit 122 acquires the binary feature of the target person from the feature conversion unit 121, and calculates the degree of agreement between the binary feature of the selected enrolled person and the binary feature of the target person. If the degree of agreement is equal to or greater than a threshold, the calculation unit 122 selects (selects) the feature linked to the binary feature of the selected enrolled person as a target for score calculation. The calculation unit 122 calculates a score between the feature linked to the binary feature of the enrolled person selected as a target for score calculation and the feature of the target person. If the degree of agreement is not equal to or greater than the threshold, the calculation unit 122 selects the next enrolled person. Note that the term "select" may be replaced with other terms such as "decide" or "set."

[0039] The calculation unit 122 sequentially selects each of the binary features of the registered persons stored in the memory unit 13, calculates the degree of match between the binary features of the selected registered person and the binary features of the subject, and calculates a score between the feature linked to the binary feature of the registered person whose degree of match is equal to or greater than a threshold and the feature of the subject.

[0040] The calculation unit 122 determines whether the highest score among the calculated scores is equal to or greater than a predetermined score, and if the highest score is equal to or greater than the predetermined score, determines that the registrant corresponding to the highest score corresponds to the target person. This identifies the registrant corresponding to the target person. The calculation unit 122 transmits information indicating the determination result to an external device via the communication unit 11.

[0041] The information processing device 1 may not need to perform a process of identifying a registrant who corresponds to the subject. In this case, the calculation unit 122 may transmit information indicating the calculated score to an external device via the communication unit 11 without identifying a registrant who corresponds to the subject.

[0042] Also, the information processing device 1 may not calculate the score. In this case, the calculation unit 122 may determine a registrant whose degree of match is equal to or greater than a threshold, and then transmit information (e.g., feature amount) about the determined registrant to an external device via the communication unit 11.

[0043] The threshold setting unit 123 sets a threshold used for comparison with the degree of coincidence in the calculation unit 122. For example, when a condition related to the setting of the threshold is instructed from an external device via the communication unit 11, the threshold setting unit 123 sets the threshold based on the instruction. A method for setting the threshold will be described later. When the threshold is provided by a user, when the threshold is fixed, etc., the information processing device 1 does not need to include the threshold setting unit 123.

[0044] The encryption processing unit 124 encrypts the feature amount and decrypts the encrypted feature amount. If encryption of the feature amount is not required, the information processing device 1 does not need to include the encryption processing unit 124.

[0045] Information related to face authentication is stored in the storage unit 13. For example, the storage unit 13 stores personal information of a registrant (identification information, name, face image, etc. of the registrant), features extracted from the face image of the registrant, and binarized features converted by the feature conversion unit 121 in association with each other.

[0046] Note that the configuration of the information processing device 1 shown in Fig. 2 is an example, and the present disclosure is not limited thereto. For example, some of the configuration shown in Fig. 2 may be omitted, or configurations not shown in Fig. 2 may be included in the information processing device 1. Also, in Fig. 2, the configurations included in the information processing device 1 may be divided into two or more devices. For example, the threshold setting unit 123 and / or the encryption processing unit 124 of the information processing device 1 may be included in another device.

[0047] <Registrant information registration> Next, an example of registration of information of a registrant will be described. When the information processing device 1 registers information of a registrant, a process of converting the feature amount of the registrant into a binary feature amount and a process of storing the converted binary feature amount are performed.

[0048] For example, when registering information on registrant #k, the feature conversion unit 121 calculates an SB sequence from the feature Pk of registrant #k, where #k is identification information for identifying the registrant, and k is an integer between 1 and N. N indicates the total number of registrants.

[0049] The method of generating the SB sequence from the feature Pk is not particularly limited. For example, if the value of the i-th element of the feature Pk is a negative value, the i-th element of the SB sequence is calculated as "1", and if the value of the i-th element of the feature Pk is not a negative value, the i-th element of the SB sequence is calculated as "0". Here, i is an integer between 1 and n, and n indicates the number of dimensions of the feature Pk. Such calculation of the elements of the SB sequence is performed for each of the 1st to nth elements of the feature Pk, and each of the 1st to nth elements of the SB sequence is calculated.

[0050] For example, when a feature is expressed in three dimensions and the feature Pk=[-10, 5, -12], the SB sequence for the feature Pk is calculated as [1, 0, 1]. Note that the SB sequence=[1, 0, 1] is hereinafter abbreviated as "101."

[0051] The feature amount conversion unit 121 associates the feature amount Pk with the SB sequence calculated from the feature amount Pk, and stores them in the storage unit 13.

[0052] 3 is a diagram showing a first example of the feature amount of a registrant and the SB sequence. As shown in FIG. 3, the storage unit 13 stores the identification information of the registrant, and the feature amount and the SB sequence corresponding to the identification information in association with each other.

[0053] When the feature amount is encrypted, the feature amount stored in FIG. 3 may be the encrypted feature amount.

[0054] Since the SB column is a binarized feature amount, the pattern of the SB column is limited. When registering the information of the registrant, the feature amount conversion unit 121 may perform grouping based on the SB column. For example, when the SB columns of a plurality of feature amounts are the same as each other, the plurality of feature amounts may be grouped as the same group.

[0055] For example, for two feature amounts Pk and Pj (j is an integer from 1 to N different from k), if the feature amount Pk = [-10, 5, -12] and the feature amount Pj = [-1, 20, -2], the SB column of the feature amount Pk and the SB column of the feature amount Pj are both "101". In this case, the feature amount Pk and the feature amount Pj (that is, registrant #k and registrant #j) belong to the same group.

[0056] FIG. 4 is a diagram showing a second example of the feature amounts and SB columns of the registrants. In FIG. 4, the respective feature amounts of the same registrants #1 to #N as in FIG. 3 are grouped based on the SB column.

[0057] As shown in FIG. 4, by performing grouping based on the SB column, the calculation of the SBM number for feature amounts with the same SB column can be omitted. For example, in the example of FIG. 4, since the SBM number for the feature amount P2 is the same as the SBM number for the feature amount P5, the calculation of the SBM number for either one of them can be omitted. Thereby, the time required for face authentication can be shortened.

[0058] Note that hereinafter, a feature amount associated with a certain SB column (a certain degree of coincidence) may be described as the feature amount of the SB column.

[0059] <Selection (Determination) of Feature Amounts Based on the SBM Number> Next, the selection of features based on the number of SBMs when face authentication is performed will be described. As described above, the calculation unit 122 of the information processing device 1 sequentially selects each of the SB strings (an example of binarized features) of the registered person stored in the storage unit 13, and calculates the number of SBMs indicating the degree of match between the SB string of the selected registered person and the SB string of the subject. Among the number of SBMs calculated in this way, the feature of the SB string corresponding to an SBM number equal to or greater than a threshold value is selected (selected) as the feature to be subjected to score calculation. Here, the selection of the feature to be subjected to score calculation may be described as narrowing down the feature to be subjected to score calculation, or pruning the feature to be subjected to score calculation.

[0060] For example, when the feature Pα of the subject α is [−10, 6, −14], the SB column of the feature Pα is “101”. In this case, the SBM number indicating the degree of agreement between the SB column of the feature Pα and the SB column “101” is “3”. Similarly, the SB column with the SBM number indicating the degree of agreement with the SB column of the feature Pα of “2” is “001”, “100”, “111”. Similarly, the SB column with the SBM number indicating the degree of agreement with the SB column of the feature Pα of “1” is “000”, “011”, “110”. Here, when the threshold value for the number of SBMs is “2”, the SB columns corresponding to the number of SBMs equal to or greater than the threshold are “101”, “001”, “100”, “111”, and therefore the features of the SB columns “101”, “001”, “100”, and “111” are selected as the features to be used for score calculation. In this case, the SB strings corresponding to the SBM counts less than the threshold, i.e., the features of "000", "011", "110", and "010", are not subject to score calculation. Score calculation is not performed for features that are not subject to score calculation. Therefore, the processing related to face recognition can be speeded up without calculating the scores for the features of each registered person, and the time required for face recognition can be shortened.

[0061] <Threshold setting> As described above, in the calculation unit 122, the SBM number indicating the degree of match between two SB strings is compared with a threshold. Hereinafter, the threshold to be compared with the SBM number is referred to as an "SBM threshold". The SBM threshold may be fixed in advance or may be determined by a predetermined method. For example, the predetermined method may be a method in which the magnitude of the score of the feature is a condition, or a method in which the order of the scores is a condition. In the following, as a setting example 1, a method of setting a threshold with the magnitude of the score of the feature as a condition will be described, and as a setting example 2, a method of setting a threshold with the order of the scores as a condition will be described.

[0062] <Setting example 1> In setting example 1, when the score between the feature of each enrolled person and the feature of the subject is calculated, an SBM threshold is set to select the feature of the enrolled person whose score with the feature of the subject is likely to be equal to or greater than the score threshold Sc. In other words, an SBM threshold is set to exclude the feature of the enrolled person who is unlikely to be equal to or greater than the score threshold Sc (or the enrolled person who is extremely unlikely to be equal to or greater than the score threshold Sc) from the score calculation target.

[0063] Fig. 5 is a diagram showing a first setting example of the SBM threshold value. Like Fig. 1, Fig. 5 shows the maximum score and the minimum score with respect to the number of SBMs.

[0064] In the example of Fig. 5, the number of SBMs whose maximum score is equal to the score threshold Sc is determined, and the determined number of SBMs is set as the SBM threshold. With this setting, the feature amount of the SBM number equal to or greater than the SBM threshold is likely to be equal to or greater than the score threshold Sc. In other words, with this setting, the feature amount of the registrant who is unlikely to be equal to or greater than the score threshold Sc (or the registrant who is unlikely to be equal to or greater than the score threshold Sc) is excluded from the score calculation target.

[0065] When setting example 1 is applied, for example, when a condition that the score threshold is equal to or greater than Sc is specified, threshold setting unit 123 has maximum score information indicating the maximum score for the number of SBMs, as illustrated in Fig. 5. When a score threshold Sc is specified, threshold setting unit 123 determines the number of SBMs whose maximum score is the score threshold Sc based on the maximum score information, and sets the determined number of SBMs as the SBM threshold.

[0066] <Setting example 2> In setting example 2, when the score between the feature of each enrolled person and the feature of the target person is calculated, an SBM threshold is set to select the feature of the enrolled person whose score is likely to be included in the top T. In other words, an SBM threshold is set to exclude the feature of the enrolled person whose score is unlikely to be included in the top T (or the enrolled person whose score is extremely unlikely to be included in the top T) from the score calculation target.

[0067] FIG. 6 is a diagram showing an example of the relationship between the number of SBMs and the number of cases. In FIG. 6, the number of SBMs, the number of cases corresponding to the number of SBMs, and the corresponding SB string are associated with each other. Here, the number of cases may be regarded as the number of features linked to the SB string corresponding to the number of SBMs, or may be regarded as the number of registered persons corresponding to the features linked to the SB string. In the example of FIG. 6, the SB string linked to the feature of the subject is "101".

[0068] Fig. 7 is a diagram showing an example of the distribution of the number of cases based on the relationship shown in Fig. 6. The horizontal axis of Fig. 7 shows the number of cases. In the example of Fig. 7, the number of cases with an SBM number of 1 or more is 22, and the number of cases with an SBM number of 2 or more is 12.

[0069] Here, when T is 10, that is, when an SBM threshold is set to select features of registered individuals expected to be included in the top 10 scores, the number of SBMs corresponding to T=10 is determined to be SBM number=2, as shown in Fig. 7. For this determined SBM number=2, the SBM threshold is set using the correspondence relationship between the maximum score and the minimum score for the SBM number.

[0070] Fig. 8 is a diagram showing a second setting example of the SBM threshold value. Like Fig. 1, Fig. 8 shows the maximum score and the minimum score with respect to the number of SBMs.

[0071] As described with reference to Fig. 7, the SBM number (in the example of Fig. 7, the SBM number = 2) corresponding to the top T (10 in the example of Fig. 7) is determined based on the distribution of the number of cases. The SBM number determined based on the distribution is referred to as a tentative SBM number. The tentative SBM number may be an example of a judgment value of the SBM number.

[0072] In setting example 2, as shown in Fig. 8, a score St is determined as the minimum score corresponding to the tentative SBM number. Next, the number of SBMs is determined with the score St being the minimum score corresponding to the tentative SBM number as the maximum score. Here, the determined number of SBMs is set as the SBM threshold. With this setting, the features of the SBM number equal to or greater than the SBM threshold may include the top T scores.

[0073] When setting example 2 is applied, for example, when a condition to select the top T scores is specified, the threshold setting unit 123 has score information indicating the minimum and maximum scores for the number of SBMs, as illustrated in Fig. 8. Then, when the top T scores are specified, the threshold setting unit 123 first determines a provisional number of SBMs corresponding to the top T scores based on the distribution of the number of SBMs (see, for example, Figs. 6 and 7). Then, the threshold setting unit 123 determines a minimum score corresponding to the provisional number of SBMs based on the minimum score information, and determines a number of SBMs with the determined minimum score as the maximum score based on the maximum score information, and sets the determined number of SBMs as the SBM threshold.

[0074] Here, the reason why setting example 2 is applied will be explained.

[0075] As described above, the greater the number of SBMs, the greater the possibility that the calculated score will be. Therefore, for example, in order to select the feature values ​​of a registrant whose score is likely to be in the top T, it is conceivable to select the feature values ​​of a registrant whose number of SBMs is in the top T as targets for score calculation. However, in this case, there is a possibility that the feature values ​​of a registrant whose number of SBMs is not in the top T but whose score is likely to be in the top T will be excluded from the targets for score calculation.

[0076] Therefore, in setting example 2, the minimum score is determined among the scores calculated from the features of enrolled persons whose SBM numbers are included in the top T, and the SBM number corresponding to the determined minimum score is set as the SBM threshold so as not to exclude enrolled persons with features that may result in a score greater than the determined minimum score.

[0077] Fig. 9 is a diagram supplementing the process of SBM threshold setting example 2. Fig. 9 shows the maximum score and minimum score for the number of SBMs similar to Fig. 8. Fig. 9 also shows two ranges, range A and range B, and two points, point Pc and point Pd, superimposed on each other.

[0078] 9, range A is the range from the maximum SBM number to the provisional SBM number, and is surrounded by the maximum score and the minimum score. As described above, the provisional SBM number corresponds to the lower limit of the SBM number in the range that includes the features of registrants with the top T SBM numbers, so range A is the range that corresponds to the features of registrants with the top T SBM numbers. Point Pc shows the relationship between the SBM number and score calculated from the features of registrant c, and point Pd shows the relationship between the SBM number and score calculated from the features of registrant d.

[0079] 9 shows a relationship in which point Pc is included in range A, while point Pd is not included in range A. This relationship corresponds to the number of SBMs of the feature of registrant c corresponding to point Pc being included in the top T, while the number of SBMs of the feature of registrant d corresponding to point Pd being not included in the top T. On the other hand, as shown in FIG. 9, since the score indicated by point Pd is higher than the score indicated by point Pc, the possibility that the score calculated from the feature of registrant d corresponding to point Pd will be included in the top T is higher than the possibility that the score calculated from the feature of registrant c corresponding to point Pc will be included in the top T.

[0080] In this way, if features falling within range A are selected as features of a registrant whose SBM count is in the top T, and features not falling within range A are not selected, the features of a registrant whose SBM count is not in the top T but whose score may be in the top T (for example, registrant d in Figure 9) will be excluded from score calculation.

[0081] Therefore, in setting example 2, the minimum value of the score calculated from the features included in range A as the features of registered users whose SBM counts are in the top T is determined, and the SBM threshold is set so that features that have the potential to calculate a score equal to or greater than the determined minimum value are included in the score calculation.

[0082] As described above, in setting example 2, a tentative SBM number corresponding to the lower limit of the number of SBMs in range A including the features of the top T registered users in terms of the number of SBMs is determined, and a minimum score St corresponding to the tentative SBM number is determined. The minimum score St corresponds to the minimum value of the scores calculated from the features included in range A.

[0083] Then, the number of SBMs with the score St as the maximum score is determined as the SBM threshold. For example, if the number of SBMs for a certain feature X is equal to or greater than this SBM threshold, the score calculated from the feature X is likely to be equal to or greater than St. In other words, if the number of SBMs for a certain feature X is less than this SBM threshold, the score calculated from the feature X is unlikely to be equal to or greater than St (or the possibility is extremely low).

[0084] The number of SBMs with St as the maximum score is determined as the SBM threshold, so that feature quantities for which scores equal to or higher than St may be calculated are included in the target for score calculation. In the example of FIG. 9, by determining the number of SBMs with St as the maximum score as the SBM threshold, it is possible to include at least the feature quantities included in range B in the target for score calculation. Note that range B is a range that is not included in range A but includes feature quantities for which the calculated scores are equal to or higher than St. That is, range B corresponds to a range that does not include the top T in terms of the number of SBMs but may include the top T in terms of scores for a registrant (in FIG. 9, for example, registrant d).

[0085] According to the two setting examples described above, an SBM threshold for selecting feature quantities expected to satisfy desired conditions can be appropriately set.

[0086] Note that in the two setting examples described above, the condition instructions may be acquired from an external device via the communication unit 11. Also, the two setting examples described above may be combined. For example, when a condition such as the score being equal to or higher than Sc and selecting the top T in terms of scores is instructed, two SBM thresholds may be set based on the two setting examples described above, or either of the two SBM thresholds may be set.

[0087] <Update of Relationship between Number of SBMs and Score> As described above, in setting the threshold, score information indicating the maximum score and / or the minimum score with respect to the number of SBMs is used. It is conceivable that the score information changes due to changes in the situation of the space where face authentication is performed. Changes in the situation where face authentication is performed are, for example, changes caused by differences in the time (daytime, evening, night, etc.) when face authentication is performed and differences in the weather (for example, sunny, cloudy, rainy, etc.) when face authentication is performed. Because such changes occur, the score information indicating the maximum score and / or the minimum score with respect to the number of SBMs may be updated. For example, the update of the score information may be executed by the threshold setting unit 123.

[0088] The score information may be updated based on the scores calculated so far. The scores calculated so far include, for example, a pair of a score calculated in face recognition score calculation and the SBM number corresponding to that score. The scores calculated so far may be scores calculated within a predetermined time period prior to the present time. The scores calculated so far are an example of history information indicating the history of score calculation.

[0089] The scores calculated so far include, for example, a score for a set of features that was excluded from the calculation target in the face recognition score calculation, and a set of the calculated score and the SBM number corresponding to that score. For example, the score for a set of features that was excluded from the calculation target in the face recognition score calculation is calculated during idle processing time (for example, time when face recognition is not performed).

[0090] The method of updating the score information is not particularly limited. The maximum score and / or the minimum score may be updated from the scores calculated so far. For example, the maximum score Smax for the SBM number Xu shown in the score information is updated to the maximum score calculated so far for the SBM number Xu. Also, for example, the minimum score Smin for the SBM number Xu shown in the score information is updated to the minimum score calculated so far for the SBM number Xu.

[0091] Also, the maximum score and / or the minimum score may be updated based on the statistical properties of the scores calculated so far. For example, the frequency of the scores calculated so far in the SBM number Xu (e.g., normal distribution assumption) is calculated, and the standard deviation σ is calculated from the calculated frequency. Then, for example, the maximum score Smax for the SBM number Xu is updated to a value that is +3 times the standard deviation σ (i.e., 3σ). Also, for example, the minimum score Smin for the SBM number Xu is updated to a value that is -3 times the standard deviation σ (i.e., -3σ).

[0092] In addition, if there is an SBM number for which the score has not been calculated (hereinafter, an uncalculated SBM number), the maximum score and minimum score for the uncalculated SBM number are not updated. The maximum score and minimum score for the uncalculated SBM number may be interpolated using the maximum score and minimum score for the other updated SBM numbers.

[0093] For example, an approximation curve showing the relationship between the SBM number and the maximum score is calculated from the updated data of the maximum scores for the other SBM numbers, and the maximum score for the uncalculated SBM number Xm is determined based on the approximation curve.The maximum score for the uncalculated SBM number Xm is then updated to the maximum score determined based on the approximation curve.Similar to the maximum score, the maximum score for the uncalculated SBM number Xm is also updated to the minimum score determined based on the approximation curve showing the relationship between the SBM number and the minimum score.

[0094] <Example of processing flow> Next, a description will be given of a process flow in the above-mentioned information processing device 1. The following describes a registration process for registering information on a registrant, a threshold setting process for setting an SBM threshold, and a score calculation process for selecting a feature amount for which a score is to be calculated and calculating a score, all of which are executed in the information processing device 1.

[0095] Fig. 10 is a flowchart showing an example of the flow of registration processing of the information processing device 1 according to the present embodiment. The flow shown in Fig. 10 is started, for example, when the information processing device 1 receives information indicating a request for registration of information of a registrant (hereinafter, a registration request) from an external device.

[0096] The communication unit 11 acquires features of a registrant from an external device (for example, a device that requests a process to register information of the registrant) (S101).

[0097] The feature amount conversion unit 121 binarizes the feature amount of the registrant (S102). By binarizing the feature amount, a binarized feature amount (for example, an SB sequence) is generated.

[0098] The encryption / decryption processing unit 124 encrypts the feature amount (S103). Note that if the information processing device 1 does not perform encryption, the process of S103 may be omitted.

[0099] The feature conversion unit 121 stores the feature and the binarized feature in the storage unit 13 (S104). When the process of S103 is executed, the feature to be stored may be an encrypted feature. Then, the flow of FIG. 10 ends.

[0100] By executing the flow in FIG. 10 for each enrolled person, the feature amount of each enrolled person is associated with the binarized feature amount and stored.

[0101] Fig. 11 is a flowchart showing an example of a flow of threshold setting in the information processing device 1 according to the present embodiment. The flow shown in Fig. 11 corresponds to a flow based on the setting example 1 described above. The flow shown in Fig. 11 is started, for example, when the information processing device 1 receives an instruction of a score threshold for determining an SBM threshold from an external device (for example, a terminal that performs face authentication). Note that, when the information processing device 1 does not determine a threshold, the flow in Fig. 11 does not need to be executed.

[0102] The communication unit 11 acquires information indicating a score threshold from an external device (for example, a device requesting face authentication) (S201).

[0103] The threshold setting unit 123 determines the degree of match (for example, the number of SBMs) for the acquired score threshold from the maximum score information (S202).

[0104] The threshold setting unit 123 sets the determined degree of match as a threshold (for example, an SBM threshold) (S203). Then, the flow shown in FIG.

[0105] Fig. 12 is a flowchart showing an example of the flow of the score calculation process of the information processing device 1 according to the present embodiment. The flow shown in Fig. 12 is started, for example, when the information processing device 1 receives information indicating a request for score calculation related to face authentication (hereinafter, score calculation request) from an external device (for example, a device requesting face authentication).

[0106] The communication unit 11 acquires features of a person to be subjected to face authentication from an external device (for example, a device requesting face authentication) (S301).

[0107] The feature amount conversion unit 121 binarizes the feature amount of the subject (S302). By binarizing the feature amount, a binarized feature amount (for example, an SB sequence) is generated.

[0108] The calculation unit 122 selects one of the unselected binary feature amounts from among the binary feature amounts of the registrants stored in the storage unit 13 (S303).

[0109] The calculation unit 122 calculates the degree of agreement (for example, the SBM number) between the binary feature amount of the subject converted in S302 and the binary feature amount of the registered person selected in S303 (S304).

[0110] The calculation unit 122 determines whether the degree of match is equal to or greater than a threshold (for example, an SBM threshold) (S305).

[0111] If the degree of match is not equal to or greater than the threshold (NO in S305), the process proceeds to S308.

[0112] If the degree of match is equal to or greater than the threshold (YES in S305), the encryption processing unit 124 decrypts the encrypted feature corresponding to the binarized feature of the registrant selected in S303 (S306). Note that if the information processing device 1 does not perform encryption, the process of S306 may be omitted.

[0113] The calculation unit 122 calculates a score between the feature of the subject acquired in S302 and the feature corresponding to the binarized feature of the enrolled person selected in S303 (S307). When the process of S306 is executed, the feature corresponding to the binarized feature of the enrolled person selected in S303 may be a decoded feature. The calculated score may be transmitted to an external device via the communication unit 11. Then, the process of S308 is executed.

[0114] The calculation unit 122 determines whether or not there is an unselected binary feature among the binary feature of the registrant stored in the storage unit 13 (S308).

[0115] If there is an unselected binarized feature (YES in S308), the process of S303 is executed again.

[0116] If there is no unselected binarized feature (NO in S308), for example, if all of the binarized feature amounts of the registered person stored in the storage unit 13 have been selected, the flow in FIG. 12 ends.

[0117] 12, when the degree of coincidence of the binary features is less than a threshold, score calculation can be omitted, thereby shortening the time required for face authentication. Also, when the encryption processing unit 124 performs encryption, when the degree of coincidence of the binary features is less than a threshold, decryption of the encrypted features can be omitted, thereby shortening the time required for face authentication.

[0118] In the present embodiment described above, the feature conversion unit 121 converts a feature acquired from a face image and included in a first numerical range into a conversion value (e.g., a binarized feature) in a second numerical range (e.g., two values) narrower than the first numerical range. The calculation unit 122 calculates an evaluation value (e.g., an SBM number) indicating a similarity between a second conversion value (e.g., an SB string of a subject) linked to a second feature of a subject of face authentication for each first conversion value (e.g., an SB string of a subject) linked to a first feature of each pre-registered registrant, and selects a first feature linked to a first conversion value having an evaluation value equal to or greater than a threshold (e.g., an SBM threshold). According to this configuration, the feature is converted into another value based on the feature, and the converted value is used to determine the feature of a registrant expected to have a high score, thereby limiting the target for score calculation to the feature of a registrant expected to have a high score. This can speed up the score calculation process and shorten the time required for face authentication.

[0119] In the present embodiment, an example of binarization has been shown as an example of converting a feature into another value based on the feature, but the present disclosure is not limited thereto. For example, the feature may be converted into a value selected from a numerical range narrower than the numerical range that the feature can take. By converting into a value selected from a numerical range narrower than the numerical range that the feature can take, the feature to be used for score calculation can be selected by processing based on the converted value, thereby reducing the time required for face recognition.

[0120] For example, the feature amount may be converted into one of four values. For example, if the feature amount is a positive value and is equal to or greater than a threshold (the threshold is a positive value), the feature amount may be converted into "2", if the feature amount is equal to or greater than 0 and less than the threshold amount, the feature amount may be converted into "1", if the feature amount is a negative value and the absolute value is equal to or greater than the threshold amount, the feature amount may be converted into "-2", and if the feature amount is a negative value and the absolute value is less than the threshold amount, the feature amount may be converted into "-1".

[0121] In this case, the product between the elements of the transformed feature is calculated as one of six values: 4, 2, 1, -1, -2, and -4. When the product between the elements of the transformed feature is calculated for each element and the products of the elements are added, the added value becomes higher according to the score between the features. Here, the added value corresponds to an example of an evaluation value indicating the similarity of the transformed feature. Here, the added value corresponds to the SBM number (or the degree of agreement) in the above-mentioned binarization example. For example, the higher the score between the features, the higher the added value may be. Therefore, similar to the SBM number in the above-mentioned embodiment, by selecting features whose added value is equal to or greater than a threshold, the target for calculating the score can be limited to the feature of the registered user whose score is expected to be high. This can speed up the score calculation process and shorten the time required for face authentication.

[0122] The present disclosure can be realized in software, hardware, or software in conjunction with hardware.

[0123] Each functional block used in the description of the above embodiments may be partially or entirely realized as an LSI, which is an integrated circuit, and each process described in the above embodiments may be partially or entirely controlled by one LSI or a combination of LSIs. The LSI may be composed of individual chips, or may be composed of one chip that includes some or all of the functional blocks. The LSI may have data input and output. Depending on the degree of integration, the LSI may be called an IC, a system LSI, a super LSI, or an ultra LSI.

[0124] The method of integration is not limited to LSI, and may be realized by a dedicated circuit, a general-purpose processor, or a dedicated processor. Also, after LSI manufacturing, a programmable FPGA (Field Programmable Gate Array) or a reconfigurable processor that can reconfigure the connections and settings of circuit cells inside the LSI may be used. The present disclosure may be realized as digital processing or analog processing.

[0125] Furthermore, if a new integrated circuit technology that can replace LSI appears due to the progress of semiconductor technology or a derivative technology, it is possible to integrate the functional blocks using that technology. The application of biotechnology is also a possibility.

[0126] The present disclosure may be implemented in any type of apparatus, device, or system having a communication function (collectively referred to as a communication apparatus). The communication apparatus may include a radio transceiver and processing / control circuitry. The radio transceiver may include a receiver and a transmitter, or both as functions. The radio transceiver (transmitter and receiver) may include a radio frequency (RF) module and one or more antennas. The RF module may include an amplifier, an RF modulator / demodulator, or the like. Non-limiting examples of communication devices include telephones (e.g., cell phones, smartphones, etc.), tablets, personal computers (PCs) (e.g., laptops, desktops, notebooks, etc.), cameras (e.g., digital still / video cameras), digital players (e.g., digital audio / video players), wearable devices (e.g., wearable cameras, smart watches, tracking devices, etc.), game consoles, digital book readers, telehealth and telemedicine devices, communication-enabled vehicles or mobile conveyances (e.g., cars, airplanes, boats, etc.), and combinations of the above devices.

[0127] Communications Equipment includes, but is not limited to portable or mobile equipment, non-portable or fixed equipment, devices and systems of any kind, such as smart home devices (appliances, lighting equipment, smart meters or metering devices, control panels, etc.), vending machines and any other "Things" that may exist on an Internet of Things (IoT) network.

[0128] In recent years, in the field of IoT (Internet of Things) technology, CPS (Cyber ​​Physical Systems) is a new concept that creates new added value by linking information between physical space and cyber space. This CPS concept can also be adopted in the above-mentioned embodiment.

[0129] That is, as a basic configuration of a CPS, for example, an edge server located in a physical space and a cloud server located in a cyberspace can be connected via a network, and processing can be distributed and processed by the processors installed in both servers. Here, each processing data generated in the edge server or the cloud server is preferably generated on a standardized platform, and the use of such a standardized platform can improve the efficiency of building a system including a variety of sensors and IoT application software.

[0130] Communications include data communications via cellular systems, wireless LAN systems, communications satellite systems, etc., as well as data communications via combinations of these.

[0131] A communications apparatus also includes devices, such as controllers and sensors, connected or coupled to a communications device that performs the communications functions described in this disclosure, such as controllers and sensors that generate control and data signals used by the communications device to perform the communications functions of the communications apparatus.

[0132] The communication apparatus also includes infrastructure facilities, such as base stations, access points, and any other apparatus, device, or system that communicates with or controls the various apparatuses listed above, but are not limited to these.

[0133] Although various embodiments have been described above with reference to the drawings, it goes without saying that the present disclosure is not limited to such examples. It is clear that a person skilled in the art can come up with various modified or amended examples within the scope of the claims, and it is understood that these also naturally belong to the technical scope of the present disclosure. In addition, the components in the above-mentioned embodiments may be arbitrarily combined within the scope of the disclosure.

[0134] Although specific examples of the present disclosure have been described in detail above, these are merely examples and do not limit the scope of the claims. The technology described in the claims includes various modifications and alterations of the specific examples exemplified above. [Industrial Applicability]

[0135] An embodiment of the present disclosure is suitable for a face recognition system. [Explanation of symbols]

[0136] 1. Information processing device 11 Communications Department 12 processors 13 Storage section 121 Feature Transformation Unit 122 Calculation section 123 Threshold determination unit 124 Encryption processing unit

Claims

1. a feature conversion unit that converts a feature acquired from a face image and falling within a first numerical range into a converted value within a second numerical range that is narrower than the first numerical range; a calculation unit that calculates an evaluation value indicating a similarity between a first conversion value associated with a second feature of a subject of face authentication for each first conversion value stored in association with a first feature of each pre-registered person, and the second conversion value associated with a second feature of the subject of face authentication, and selects a first feature associated with a first conversion value for which the evaluation value is equal to or greater than a threshold; An information processing device comprising:

2. a threshold setting unit that determines the threshold based on a statistical relationship between the evaluation value between two converted values ​​obtained by converting each of the two feature amounts and a score indicating a similarity between the two feature amounts; Further preparations: The information processing device according to claim 1 .

3. the statistical relationship indicates a maximum value of the score for the evaluation value; When a condition that the score is equal to or greater than S is specified, the threshold setting unit determines an evaluation value whose maximum value is S in the statistical relationship, and sets the determined evaluation value as the threshold. The information processing device according to claim 2 .

4. the threshold setting unit updates the statistical relationship based on history information having a pair of the score calculated by the calculation unit and the evaluation value between the two feature amounts that are the subject of calculation of the score; 4. The information processing device according to claim 2 or 3.

5. The threshold setting unit creates an approximation curve based on the history information, updating the maximum value of the score included in the statistical relationship based on the approximation curve; The information processing device according to claim 4.

6. the statistical relationship indicates a maximum value of the score and a minimum value of the score for the evaluation value, The threshold setting unit is When a condition that the top T scores are selected is specified, the feature amounts of the registered person are selected in order from the feature amount with the largest evaluation value, and a judgment value of the evaluation value including at least the T scores is determined; A minimum value of the scores where the evaluation value corresponds to the judgment value in the statistical relationship is set as a provisional score; determining an evaluation value where the maximum value of the scores corresponds to the provisional score; setting the determined evaluation value to the threshold value; The information processing device according to claim 2 .

7. the threshold setting unit updates the statistical relationship based on history information having a pair of the score calculated by the calculation unit and the evaluation value between the two feature amounts that are the subject of calculation of the score; The information processing device according to claim 6.

8. The threshold setting unit creates an approximation curve based on the history information, updating a maximum value and / or a minimum value of the score included in the statistical relationship based on the approximation curve; The information processing device according to claim 7.

9. an encryption processing unit that encrypts the feature amount and decrypts the encrypted feature amount; the encryption processing unit encrypts the first feature amount acquired from each of the enrolled persons, and decrypts the first feature amount determined by the calculation unit. The information processing device according to claim 1 .

10. An information processing device, converting a feature amount obtained from a face image and included in a first numerical range into a conversion value within a second numerical range narrower than the first numerical range; calculating an evaluation value indicating a similarity between a first conversion value associated with a first feature amount of each pre-registered person and a second conversion value associated with a second feature amount of a subject of face authentication, and selecting a first feature associated with a first conversion value having an evaluation value equal to or greater than a threshold; Information processing methods.

11. In the information processing device, converting a feature amount obtained from a face image and included in a first numerical range into a conversion value within a second numerical range narrower than the first numerical range; calculating an evaluation value indicating a similarity between a first conversion value associated with a first feature amount of each pre-registered person and a second conversion value associated with a second feature amount of a subject of face authentication, and selecting a first feature associated with a first conversion value having an evaluation value equal to or greater than a threshold; A program that executes a process.

Images