Face authentication device, face authentication method, and program

Abstract

To provide a face authentication device which can quickly perform collation with a large volume of collation objects by means of distributed processing by using a proper key.SOLUTION: A face authentication device 50 comprises: a plurality of collation means which store a plurality of face feature amounts corresponding to a plurality of faces in a distributed manner on the basis of a value of a shard key; acquisition means; shard key calculation means; and authentication means. The acquisition means acquires an input face image. The shard key calculation means extracts a plurality of line segments respectively determined by a feature point of the face from the input face image and calculates a ratio of the plurality of line segments as the shard key. The authentication means performs authentication of the input face image by using the collation means corresponding to the value of the calculated shard key.SELECTED DRAWING: Figure 10

Description

Technical Field

[0001] The present invention relates to face recognition technology.

Background Art

[0002] As a method for efficiently searching a large amount of data, a technique of dividing a search target by sharding is known. For example, Patent Document 1 describes a method of calculating a weighting value indicating the degree of collation for each combination of a collation source and a collation destination in biometric authentication, and using this to classify the feature vectors of the collation destination into a plurality of groups.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Even in face authentication, when the number of collation targets becomes enormous, it is expected to speed up the authentication process by sharding. However, in the distributed analysis of unstructured data such as face authentication, the lack of an appropriate key that can be used for sorting in the pre-collation stage has been an issue.

[0005] One object of the present invention is to provide a face authentication device capable of performing collation with a large number of collation targets at high speed by distributed processing using an appropriate key.

Means for Solving the Problems

[0006] In one aspect of the present invention, a face authentication device acquisition means for acquiring an input face image, From the input face image, a plurality of line segments each defined by a facial feature point are extracted, and the complex shard key calculation means for calculating the ratio of a number of line segments as a shard key; authentication means for authenticating the input face image using a pre-registered feature amount corresponding to the value of the calculated shard key and the feature amount of the input face image; It is provided with.

[0007] In another aspect of the present invention, the face authentication method is executed by a computer, acquire an input face image, extract a plurality of line segments each defined by face feature points from the input face image, and calculate the ratio of the plurality of line segments as a shard key, calculate the ratio of the plurality of line segments as a shard key, authenticate the input face image using a pre-registered feature amount corresponding to the value of the calculated shard key and the feature amount of the input face image.

[0008] In still another aspect of the present invention, the program acquire an input face image, extract a plurality of line segments each defined by face feature points from the input face image, and calculate the ratio of the plurality of line segments as a shard key, calculate the ratio of the plurality of line segments as a shard key, cause a computer to execute a process of authenticating the input face image using a pre-registered feature amount corresponding to the value of the calculated shard key and the feature amount of the input face image.

Brief Description of Drawings

[0009]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

[0010] Hereinafter, with reference to the drawings, preferred embodiments of this disclosure will be described. <First Embodiment> [Face Authentication System] FIG. 1 shows an overview of the face authentication system according to the first embodiment. The face authentication system 1 includes a terminal device 5 and a face authentication server 100. The terminal device 5 is a client terminal used by a user who performs face authentication, and examples thereof include the user's PC, tablet, smartphone, etc. The user transmits a face image captured by a camera or the like from the terminal device 5 to the face authentication server 100.

[0011] The face authentication server 100 stores in advance face images and feature amounts of face images for registered persons, and collates the face image to be authenticated (hereinafter, also referred to as "input face image") transmitted from the terminal device 5 with the registered face images to perform face authentication. Specifically, the face authentication server 10 extracts feature amounts from the input face image and collates them with a plurality of feature amounts registered in the face authentication server 10 to perform face authentication.

[0012] [Hardware Configuration] FIG. 2 is a block diagram showing the hardware configuration of the face authentication server 100. As shown in the figure, the face authentication server 100 includes a communication unit 11, a processor 12, a memory 13, a recording medium 14, and a database (DB) 15.

[0013] The communication unit 11 performs input / output of data with an external device. Specifically, the communication unit 11 receives an input face image to be authenticated from the terminal device 5. Further, the communication unit 11 transmits the authentication result by the face authentication server 100 to the terminal device 5.

[0014] The processor 12 is a computer such as a CPU (Central Processing Unit), and controls the entire face authentication server 100 by executing a program prepared in advance. Note that the processor 12 may be a GPU (Graphics Processing Unit) or an FPGA (Field-Programmable Gate Array). Specifically, the processor 12 executes face authentication processing described later.

[0015] The memory 13 is composed of a ROM (Read Only Memory), a RAM (Random Access Memory), etc. The memory 13 is also used as a working memory during the execution of various processes by the processor 12.

[0016] The recording medium 14 is a non-volatile and non-temporary recording medium such as a disk-shaped recording medium or a semiconductor memory, and is configured to be detachable from the face authentication server 100. The recording medium 14 stores various programs executed by the processor 12. When the face authentication server 100 executes face authentication processing, the programs recorded on the recording medium 14 are loaded into the memory 13 and executed by the processor 12.

[0017] The database (DB) 15 stores the face images and the feature amounts extracted from the face images (hereinafter also referred to as "registered feature amounts") for the registered persons. Further, the DB 15 temporarily stores the input face images received through the communication unit 11, the authentication results of the input face images, and the like. In addition, the face authentication server 100 may include a display unit, an input unit, etc. for the administrator to perform necessary operations and the like.

[0018] [Functional Configuration] FIG. 3 is a block diagram showing the functional configuration of the face authentication server 100. Functionally, the face authentication server 100 includes a data acquisition unit 20, a calculation unit 21, a determination unit 22, face matching nodes 23a to 23d, a full feature amount DB 25, and an authentication result output unit 26.

[0019] The data acquisition unit 20 acquires the input face image to be authenticated from the user's terminal device 5 and outputs it to the calculation unit 21. The calculation unit 21 extracts the feature amounts (hereinafter also referred to as "feature amounts for comparison") for comparison with the registered face images from the input face image, and outputs a predetermined feature amount in the input face image as a shard key to the determination unit 22.

[0020] The calculation unit 21 uses, as the shard key, the feature amounts that can be relatively easily acquired from the information obtained from the face image and that have stable values. There is a lot of ambiguous information obtained from the face image. For example, the age and gender estimated from the face image have a large decrease in determination accuracy due to makeup, wigs, clothing, etc. Therefore, if these are used as the shard key, it is considered that the false determination increases. Therefore, in the present embodiment, the aspect ratio of the T shape of both eyes and the upper lip is used as the shard key.

[0021] Figure 4 shows the method for calculating the shard key in this embodiment. As shown in the figure, the aspect ratio of the T shape formed by both eyes and the upper lip of the face is used as the shard key. Specifically, the arithmetic unit 21 determines a first line segment connecting both eyes in the input face image, and further uses the perpendicular line dropped from the upper lip to the first line segment as the second line segment, and sets the ratio (W / H) of the length (W) of the first line segment and the length (H) of the second line segment as the shard key. Note that the arithmetic unit 21 calculates the shard key after performing necessary corrections on the input face image in the pan, roll, and tilt directions. The above ratio (W / H), that is, the aspect ratio of the T shape of both eyes and the upper lip, can be easily measured from the face image and can be stably measured without depending on makeup or the like, so the possibility of misjudgment can be reduced, and stable distributed processing of face authentication becomes possible.

[0022] The all-feature DB 25 is a DB that stores all registered feature amounts of registered persons, and is realized by the DB 15 in FIG. 2. The all-feature amount data stored in the all-feature DB 25 is distributed to a plurality of nodes by sharding. In the example of FIG. 3, the all-feature amount data is distributed and stored in four face matching nodes 23a to 23d based on the shard key. The face matching nodes 23a to 23d are provided with cache memories 24a to 24d for storing feature amount data. In the following description, when referring to any one of the four face matching nodes or cache memories, a subscript such as "face matching node 23a" is added, and when not limited to any one, the subscript is omitted and it is called "face matching node 23" etc.

[0023] Figure 5 shows a method for allocating all-feature amount data to four face matching nodes 23 using the shard key. This process is executed by the arithmetic unit 21 and the determination unit 22. First, for all registered face images, the aspect ratio of the T shape of both eyes and the upper lip is calculated as the shard key. Thereby, the value of the shard key corresponding to the all-feature amount data is obtained. Next, as shown in FIG. 5, the all-feature amount data is sorted by the value of the shard key. Then, the all-feature amount data is classified into four groups G1 to G4 by the value of the shard key and allocated to the face matching nodes 23a to 23d.

[0024] Specifically, the group G1 of feature data with the largest shard key value is stored in the cache memory 24a of the face matching node 23a, and the group G2 of feature data with the second largest shard key value is stored in the cache memory 24b of the face matching node 23b. Similarly, the group G3 of feature data with the third largest shard key value is stored in the cache memory 24c of the face matching node 23c, and the group G4 of feature data with the smallest shard key value is stored in the cache memory 24d of the face matching node 23d. In this way, each cache memory 24a to 24d stores the feature data of each group grouped into four by the shard key value. As a result, it is not necessary to query the entire feature amount DB25 in which all face feature amounts are stored every time the matching described later is performed, so that the processing speed can be increased.

[0025] When allocating all the feature data to the four face matching nodes 23a to 23d, considering that there may be some errors in the shard key values calculated from the face images, it is preferable to slightly overlap the boundaries of adjacent groups as shown in FIG. 5. In the example of FIG. 5, the feature data included in the region X1 is stored in both the face matching nodes 23a and 23b, the feature data included in the region X2 is stored in both the face matching nodes 23b and 23c, and the feature data included in the region X3 is stored in both the face matching nodes 23c and 23d. Note that FIG. 3 is merely an example, and the number of face matching nodes 23 can be determined according to the number of all feature data.

[0026] As described above, the process of allocating all the feature data to the four face matching nodes 23 is performed before starting the face authentication process for the input face image. Then, when the allocation is completed, the value range of the shard key corresponding to the group of feature data allocated to each face matching node 23a to 23d is stored in the determination unit 22.

[0027] When the determination unit 22 obtains the value of the shard key calculated for the input face image from the calculation unit 21, it selects the face matching node 23 to which the value belongs, and outputs the feature amount for face matching of the input face image to the selected face matching node 23. For example, if the value of the shard key obtained from the calculation unit 21 for a certain input face image is the value corresponding to the face matching node 23b, the determination unit 22 outputs the feature amount for face matching of the input face image to the face matching node 23b.

[0028] The face matching node 23 that receives the feature amount for face matching of the input face image from the determination unit 22 compares the input feature amount for face matching with the feature amount data stored in the cache memory 24 to perform face authentication. For example, if the determination unit 22 outputs the feature amount for face matching of the input face image to the face matching node 23b based on the value of the shard key, the face matching node 23b performs face authentication using the feature amount for face matching. Basically, the other three face matching nodes 23a, 23c, and 23d do not perform face authentication. When the value of the shard key of the input face image belongs to any of the regions X1 to X3 shown in FIG. 5, two face matching nodes 23 adjacent to any of the regions X1 to X3 perform face authentication, but the remaining two face authentication nodes 23 do not perform face authentication. In this way, the matching of the feature amounts of the input face image can be accelerated by distributed processing.

[0029] The face matching node 23 that has performed face authentication using the feature amount for face matching of the input face image outputs the authentication result to the authentication result output unit 26. The authentication result output unit 26 transmits the authentication result to the user's terminal device 5. In this way, the user can know the authentication result.

[0030] In the above configuration, the data acquisition unit 20 is an example of acquisition means, the calculation unit 21 is an example of shard key calculation means and feature amount extraction means, and the determination unit 22 and the face matching node 23 are examples of authentication means. Also, the calculation unit 21 and the determination unit 22 are examples of storage control means, and the all-feature amount DB 25 is an example of storage means.

[0031] [Face Authentication Process] Next, the face authentication process performed by the face authentication server 100 will be described. FIG. 6 is a flowchart of the face authentication process. This process is realized by the processor 12 shown in FIG. 2 executing a program prepared in advance and operating as an element shown in FIG. 3. As a premise of the process, it is assumed that all feature amount data is distributed and cached in the face matching nodes 23a to 23d based on the value of the shard key as described above.

[0032] First, the data acquisition unit 20 acquires an input face image from the terminal device 5 (step S11). Next, the calculation unit 21 calculates a feature amount for verification from the input face image (step S12), and further calculates a shard key from the input face image (step S13).

[0033] Next, the determination unit 22 selects the face matching node 23 corresponding to the calculated shard key from among the plurality of face matching nodes 23, and outputs the feature amount for verification to the selected face matching node 23 (step S14). The selected face matching node 23 collates the feature amount for verification of the input face image calculated in step S12 with the feature amount cached in the selected face matching node 23 (step S15). Then, the result output unit 26 outputs the collation result to the terminal device 5 (step S16). Thus, the face authentication process ends.

[0034] As described above, in the first embodiment, since the aspect ratio of the T shape of both eyes and the upper lip is used as the shard key, the shard key can be easily and stably calculated from the face image, and the distributed processing of face authentication can be stably performed.

[0035] <Second Embodiment> Next, the second embodiment will be described. In the first embodiment, the aspect ratio of the T shape of both eyes and the upper lip, specifically, the ratio (W / H) of the length (W) of the first line segment connecting both eyes in the face image and the length (H) of the second line segment which is a perpendicular line dropped from the upper lip to the first line segment is used as the shard key. In contrast, in the second embodiment, the following line segment lengths (distances) that can be stably acquired from the face image are used.

[0036] (1) The length of the line segment connecting the ear hole on either the left or right side and the pupil center (2) The length of the line segment connecting the two eyes (3) The length of the perpendicular line dropped from the upper lip to the line segment connecting the two eyes (4) The length of the perpendicular line dropped from the tip of the nose to the line segment connecting the two eyes (5) The length of the perpendicular line dropped from the tip of the chin to the line segment connecting the two eyes (6) The length of the line segment connecting the tip of the nose and the tip of the chin (7) The length of the line segment connecting the lower lip and the tip of the chin (8) The length of the line segment connecting the ear hole on either the left or right side and the tip of the nose (9) The length of the perpendicular line dropped from the eyes to the line segment connecting the ear hole on either the left or right side and the tip of the nose

[0037] Specifically, in the second embodiment, a ratio of two or more of the above-mentioned lengths is used as a shard key. Thereby, a shard key can be generated using a ratio of lengths that can be obtained relatively stably from a face image. Except for this point, the second embodiment is the same as the first embodiment.

[0038] <The Third Embodiment> Next, the third embodiment will be described. In the third embodiment, based on the pan, roll, and tilt amounts of the input face image, it is determined which part of the face is used to calculate the shard key. Specifically, the arithmetic unit 21 calculates the pan, roll, and tilt from the input face image, and refers to the first table shown in FIG. 7 to determine the distance (length of the line segment) to be used as the shard key. FIG. 7 shows the correspondence between the pan, roll, and tilt ranges of the input face image and the distance (length of the line segment) to be used in that range. In the first table of FIG. 7, the pan is set to 0 degrees for the front direction of the user's face, with negative values for the left direction and positive values for the right direction. Also, the tilt is set to 0 degrees for the front direction of the user's face, with positive values for the upward direction and negative values for the downward direction.

[0039] For example, as a result of the arithmetic unit 21 calculating the pan and tilt from the input face image, if the pan belongs to “-20° to 20°” and the tilt belongs to the range of “-30° to 30°”, the distance (2), that is, the line segment connecting both eyes, can be used. At the same time, if the pan belongs to “-20° to 20°” and the tilt belongs to the range of “-60° to 60°”, the distance (3), that is, the perpendicular line dropped from the upper lip to the line segment connecting both eyes, can be used. Therefore, the arithmetic unit 21 can use the ratio of the lengths of these two line segments as the shard key. When generating the shard key using the distances selected as described above, the number of face matching nodes 23 may be increased as necessary.

[0040] According to the third embodiment, based on the result of calculating the pan and tilt from the input face image, the shard key is calculated using the facial parts of the face shown in the input face image, so that the distributed processing by the shard key can be stabilized. Except for this point, the third embodiment is the same as the first embodiment.

[0041] <Fourth Embodiment> Next, the fourth embodiment will be described. In the fourth embodiment, based on whether the input face image is an image with a mask or sunglasses worn, it is determined which facial parts are used to calculate the shard key. Specifically, the arithmetic unit 21 analyzes the input face image to detect the mask and sunglasses, and refers to the second table shown in FIG. 8 to determine the distance (length of the line segment) used as the shard key. FIG. 8 shows the correspondence between the presence or absence of the mask and sunglasses and the distance (length of the line segment) to be used as the shard key.

[0042] For example, if the arithmetic unit 21 detects a mask from the input face image but does not detect sunglasses, the ratio of two or more of the distances (1), (2), and (9) can be used as the shard key. When generating the shard key using the distances selected as described above, the number of face matching nodes 23 may be increased as necessary.

[0043] According to the fourth embodiment, based on the result of detecting the mask and sunglasses from the input face image, a shard key is calculated using the facial parts in the input face image, so that the distributed processing by the shard key can be stabilized. Except for this point, the fourth embodiment is the same as the first embodiment.

[0044] <Fifth Embodiment> Next, the fifth embodiment will be described. In the fifth embodiment, a plurality of line segments defined by the parts included in the face image are used for calculating the shard key according to a predetermined priority order. FIG. 9 is an example of a third table defining the priority order for a plurality of line segments defined by the parts included in the face image. The arithmetic unit 21 detects the facial parts in the input face image, and calculates the shard key using the lengths of the available line segments according to the priority order shown in FIG. 9.

[0045] Now, assume that the ratio of the lengths of two line segments is used as the shard key. For example, when all the facial parts are shown in the input face image, the arithmetic unit 21 calculates the ratio of distance (2) and distance (3) as the shard key according to the priority order shown in FIG. 9. On the other hand, when the vicinity of the eyes is not shown in the input face image, the arithmetic unit 21 calculates the ratio of distance (6) and distance (7) that does not use the position of the eyes as the shard key according to the priority order shown in FIG. 9. When generating the shard key using the distances selected as above, the number of face matching nodes 23 may be increased as necessary.

[0046] According to the fifth embodiment, since the distance is selected according to the priority order from a plurality of facial parts in the input face image to calculate the shard key, the distributed processing by the shard key can be stabilized. Except for this point, the fifth embodiment is the same as the first embodiment.

[0047] <Sixth Embodiment> Next, the sixth embodiment will be described. FIG. 10 is a block diagram showing the functional configuration of a face authentication device 50 according to the sixth embodiment. The face authentication device 50 includes a plurality of collation means 51, an acquisition means 52, a shard key calculation means 53, and an authentication means 54. The plurality of collation means 51 stores, by sorting, a plurality of face feature amounts corresponding to a plurality of faces based on the values of shard keys. The acquisition means 52 acquires an input face image. The shard key calculation means 53 extracts a plurality of line segments each defined by face feature points from the input face image, and calculates the ratio of the plurality of line segments as a shard key. The authentication means 54 authenticates the input face image using the collation means corresponding to the calculated shard key value.

[0048] FIG. 11 is a flowchart of the face authentication process according to the sixth embodiment. First, the acquisition means 52 acquires an input face image (step S31). Next, the shard key calculation means 53 extracts a plurality of line segments each defined by face feature points from the input face image, and calculates the ratio of the plurality of line segments as a shard key (step S32). Then, the authentication means 54 authenticates the input face image using the collation means corresponding to the calculated shard key value among the plurality of collation means that store, by sorting, a plurality of face feature amounts corresponding to a plurality of faces based on the values of shard keys (step S33).

[0049] According to the sixth embodiment, since a plurality of line segments each defined by face feature points are extracted from the input face image and the ratio of the plurality of line segments is used as a shard key, a shard key can be stably generated from the face image, and distributed processing of face authentication can be stably performed.

[0050] <Seventh Embodiment> At a certain point in time, after the registered feature amounts stored in the all-feature amount DB 25 are distributed and stored in a plurality of face collation nodes 23, if the number of registered persons increases, there may be a bias in the number of registered feature amounts stored in each face collation node 23. In that case, a process of redistributing the registered feature amounts stored in the all-feature amount DB 25 to the plurality of face collation nodes 23, that is, an update process, may be performed.

[0051] Also, as the number of registered persons increases, the number of face matching nodes 23 may be increased. For example, assume that at a certain point in time, the registered feature amounts are distributed and stored among four face matching nodes 23 as illustrated in FIG. 3. After that, if the number of registered persons increases and the total number of registered feature amounts in the entire feature amount DB 25 exceeds a predetermined reference number, one face matching node 23 may be added, and all the registered feature amounts at that time may be redistributed among the five face matching nodes 23.

[0052] <Eighth Embodiment> In the above-described first to seventh embodiments, the method of this disclosure is applied to face authentication. However, it can also be applied to biometric authentication related to the hand, such as fingerprint authentication, hand vein authentication, and palmprint authentication. Note that for hand vein authentication and palmprint authentication, it is assumed that a captured image of the entire hand can be obtained. In this case, for example, by using, as a shard key, the ratio of the distances from the base to the tip of each of the five fingers, or the ratio of the length between the first joint and the second joint of a specific finger to the length of the finger, the same processing as in the above-described embodiments can be executed.

[0053] Some or all of the above-described embodiments may be described as follows in the following supplementary notes, but are not limited thereto.

[0054] (Supplementary Note 1) A plurality of matching means for respectively distributing and storing a plurality of face feature amounts corresponding to a plurality of faces based on the values of shard keys; An acquisition means for acquiring an input face image; Shard key calculation means for extracting a plurality of line segments each defined by face feature points from the input face image and calculating the ratio of the plurality of line segments as a shard key; Authentication means for authenticating the input face image by using the matching means corresponding to the value of the calculated shard key; A face authentication device comprising the same.

[0055] (Supplementary Note 2) The face authentication device according to Supplementary Note 1, wherein the shard key is a ratio of the length of a line segment connecting both eyes in a face image to the length of a perpendicular line dropped from the upper lip to the line segment.

[0056] (Supplementary Note 3) The line segment is any one of a line segment connecting the hole of either the left or right ear and the pupil center, a line segment connecting both eyes, a perpendicular line dropped from the upper lip to the line segment connecting both eyes, a perpendicular line dropped from the tip of the nose to the line segment connecting both eyes, a perpendicular line dropped from the tip of the chin to the line segment connecting both eyes, a line segment connecting the tip of the nose and the tip of the chin, a line segment connecting the lower lip and the tip of the chin, a line segment connecting the hole of either the left or right ear and the tip of the nose, and a perpendicular line dropped from the eyes to the line segment connecting the hole of either the left or right ear and the tip of the nose. The face authentication device according to Supplementary Note 1.

[0057] (Supplementary Note 4) It includes a first table that defines the lengths of line segments to be used in calculating the shard key corresponding to the amounts of pan, roll, and tilt of a face image. The shard key calculation means calculates the amounts of pan, roll, and tilt of the input face image, determines the line segment to be used by referring to the first table, and calculates the shard key. The face authentication device according to any one of Supplementary Notes 1 to 3.

[0058] (Supplementary Note 5) It includes a second table that defines the lengths of line segments to be used as the shard key when a face image includes at least one of a mask and sunglasses. The shard key calculation means detects whether the input face image includes a mask and sunglasses, determines the line segment to be used by referring to the second table, and calculates the shard key. The face authentication device according to any one of Supplementary Notes 1 to 3.

[0059] (Supplementary Note 6) It includes a third table that shows the priority order of a plurality of line segments defined by face feature points. The shard key calculation means selects a plurality of line segments with a higher priority order from a plurality of line segments defined by the feature points included in the input face image and calculates the shard key. The face authentication device according to any one of Supplementary Notes 1 to 3.

[0060] (Appendix 7) It is provided with feature quantity extraction means for extracting feature quantities from the input face image, The authentication means is the face authentication device according to any one of Appendices 1 to 6 that collates the feature quantity extracted from the input face image with the feature quantity stored in the collation means.

[0061] (Appendix 8) The face authentication device according to any one of Appendices 1 to 7, further comprising storage control means for distributing and storing the feature quantities of the plurality of faces stored in the storage means to the plurality of collation means based on the shard key.

[0062] (Appendix 9) Obtain an input face image, Extract a plurality of line segments each defined by face feature points from the input face image, calculate the ratio of the plurality of line segments as a shard key, Among the plurality of collation means that respectively distribute and store a plurality of face feature quantities corresponding to a plurality of faces based on the value of the shard key, use the collation means corresponding to the calculated value of the shard key to authenticate the input face image. A face authentication method.

[0063] (Appendix 10) Obtain an input face image, Extract a plurality of line segments each defined by face feature points from the input face image, calculate the ratio of the plurality of line segments as a shard key, Among the plurality of collation means that respectively distribute and store a plurality of face feature quantities corresponding to a plurality of faces based on the value of the shard key, use the collation means corresponding to the calculated value of the shard key to authenticate the input face image. A recording medium recording a program for causing a computer to execute the process.

[0064] As described above, the present invention has been described with reference to the embodiments and examples, but the present invention is not limited to the above embodiments and examples. Various changes that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention.

Explanation of Signs

[0065] 5 Terminal device 12 Processor 20 Data acquisition unit 21 Arithmetic unit 22 Judgment unit 23 Face matching node 24 Cache memory 25 All feature amount DB 26 Result output unit 100 Face authentication server

Claims

1. An acquisition means for acquiring an input face image; From the input face image, a plurality of line segments each defined by facial feature points are extracted, and a shard key calculation means for calculating the ratio of the plurality of line segments as a shard key; Using a pre-registered feature amount corresponding to the value of the calculated shard key and the feature amount of the input face image, an authentication means for authenticating the input face image; A face authentication device comprising:

2. The face authentication device according to claim 1, wherein the shard key is a ratio of the length of a line segment connecting both eyes in a face image and the length of a perpendicular line dropped from the upper lip to the line segment.

3. The line segment is any one of a line segment connecting the hole of either the left or right ear and the pupil center, a line segment connecting both eyes, a perpendicular line dropped from the upper lip to the line segment connecting both eyes, a perpendicular line dropped from the tip of the nose to the line segment connecting both eyes, a perpendicular line dropped from the tip of the chin to the line segment connecting both eyes, a line segment connecting the tip of the nose and the tip of the chin, a line segment connecting the lower lip and the tip of the chin, a line segment connecting the hole of either the left or right ear and the tip of the nose, and a perpendicular line dropped from the eye to the line segment connecting the hole of either the left or right ear and the tip of the nose. The face authentication device according to claim 1.

4. Comprising a first table that defines the lengths of line segments to be used in calculating the shard key corresponding to the amounts of pan, roll, and tilt of the face image, The shard key calculation means calculates the amounts of pan, roll, and tilt of the input face image, determines the line segments to be used with reference to the first table, and calculates the shard key. The face authentication device according to any one of claims 1 to 3.

5. Comprising a second table that defines the lengths of line segments to be used as the shard key when the face image includes at least one of a mask and sunglasses, The shard key calculation means detects whether the input face image includes a mask and sunglasses, determines the line segments to be used with reference to the second table, and calculates the shard key. The face authentication device according to any one of claims 1 to 3.

6. Comprising a third table indicating the priority order of a plurality of line segments defined by facial feature points, The shard key calculation means selects a plurality of line segments with a higher priority from the plurality of line segments defined by the feature points included in the input face image, and calculates the shard key. The face authentication device according to any one of claims 1 to 3.

7. The plurality of pre-registered feature amounts are registered in a distributed manner to a plurality of face matching nodes based on the shard key. The face authentication device according to any one of claims 1 to 6. ​

8. A feature quantity extraction means for extracting a feature quantity from the input face image is provided. The authentication means is the face authentication device according to claim 7, which collates the feature quantity extracted from the input face image with the feature quantity registered in advance in the face collation node corresponding to the calculated shard key.

9. Executed by a computer, acquires an input face image, extracts a plurality of line segments each defined by facial feature points from the input face image, and calculates the ratio of the plurality of line segments as a shard key, A face authentication method for authenticating the input face image by using the pre-registered feature quantity corresponding to the value of the calculated shard key and the feature quantity of the input face image.

10. acquires an input face image, extracts a plurality of line segments each defined by facial feature points from the input face image, and calculates the ratio of the plurality of line segments as a shard key, A program that causes a computer to execute a process of authenticating the input face image by using the pre-registered feature quantity corresponding to the value of the calculated shard key and the feature quantity of the input face image.

Images