Biometric determination device, biometric determination method, and recording medium

Abstract

The present invention comprises: an occluding region determination unit 211 that determines whether or not a face region FR including the face of a subject that is shown in an image IMG generated by photographing the subject includes an occluding region MR that occludes at least a partial region of the face; a first determination unit 212 that determines whether or not the subject is a biological object by a first determination method using the image IMG when the occluding region determination unit 211 has determined that the face region FR does not include the occluding region MR; and a second determination unit 213 that determines whether or not the subject is a biological object by a second determination method that uses the image IMG and is different from the first determination method when the occluding region determination unit 211 has determined that the face region FR includes the occluding region MR.

Description

Technical Field

[0001] This disclosure relates to the technical fields of a biological authentication device, a biological authentication method, and a recording medium.

Background Art

[0002] A user who performs face recognition can select by himself / herself the facial part to be the object for extracting feature amounts. Thus, even when wearing a mask or an eye bandage, or having a changed hairstyle, the feature amounts of the selected part are emphasized and the feature amounts of other parts are reduced, and a technique for appropriately performing face recognition is described in Patent Document 1. For an object represented by an image obtained by imaging, a likelihood value p1 indicating object-likeness is calculated using a machine learning system learned to react to an image of an object wearing a mask and an image of an object not wearing a mask, a likelihood value p2 indicating object-likeness of an object wearing a mask is calculated using a machine learning system learned to react to an image of an object wearing a mask, a likelihood value p3 indicating object-likeness of an object not wearing a mask is calculated using a machine learning system learned to react to an image of an object not wearing a mask, and based on the three calculated likelihood values p1, p2, and p3, a technique for determining whether the object represented by the imaged obtained image is wearing a mask or not is described in Patent Document 2. A technique including a step of detecting a face area from an input video in which an input face is shown, a step of generating a composite video by synthesizing a face video and a reference video based on a masking area, a step of determining a face feature corresponding to the input face based on the face video and the composite video, and a step of determining whether authentication is successful or not based on the determined face feature is described in Patent Document 3.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Patent Document 2

Patent Document 3

SUMMARY OF THE INVENTION

PROBLEMS TO BE SOLVED BY THE INVENTION

[0004] This disclosure aims to provide a biological determination device, a biological determination method, and a recording medium for improving the technology described in prior art documents.

MEANS FOR SOLVING THE PROBLEMS

[0005] One aspect of the biological determination device includes a shielding area determination means for determining whether at least a part of the face area including the face of the subject in the image is shielded, a first determination means for determining whether the subject is a living body by a first determination method using the image when the face area is not shielded, and a second determination means for determining whether the subject is a living body by a second determination method different from the first determination method using the image when the face area is shielded.

[0006] One aspect of the biological determination method is to determine whether at least a part of the face area including the face of the subject in the image is shielded, determine whether the subject is a living body by a first determination method using the image when the face area is not shielded, and determine whether the subject is a living body by a second determination method different from the first determination method using the image when the face area is shielded.

[0007] One aspect of the recording medium records a computer program for causing a computer to execute a biological determination method of determining whether at least a part of the face area including the face of the subject in the image is shielded, determining whether the subject is a living body by a first determination method using the image when the face area is not shielded, and determining whether the subject is a living body by a second determination method different from the first determination method using the image when the face area is shielded.

BRIEF DESCRIPTION OF THE DRAWINGS

[0008]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

[0009] Hereinafter, embodiments of a biological determination device, a biological determination method, and a recording medium will be described with reference to the drawings. [1: First Embodiment]

[0010] First, a first embodiment of a biological determination device, a biological determination method, and a recording medium will be described. Hereinafter, the first embodiment of the biological determination device, the biological determination method, and the recording medium will be described using the biological determination device 1 to which the first embodiment of the biological determination device, the biological determination method, and the recording medium is applied. [1-1: Configuration of Biological Determination Device 1]

[0011] FIG. 1 is a block diagram showing the configuration of the biological determination device 1 in the first embodiment. As shown in FIG. 1, the biological determination device 1 includes a shielding region determination unit 11, a first determination unit 12, and a second determination unit 13. The biological determination device 1 may perform biological determination using information that can be obtained from a face region including the face of the subject that is captured in an image generated by imaging the subject. The biological determination device 1 may determine whether the subject is a living body or a forged object of another person who is impersonating using information that can be obtained from the face region. Forged objects of another person may include flat-shaped objects such as a photo in which another person is captured, a display on which another person is displayed, and three-dimensional objects such as a doll that imitates another person, a face mask such as silicon. The biological determination performed by the biological determination device 1 may be a liveness determination for determining whether the person is actually being imaged.

[0012] The shielding region determination unit 11 determines whether a face region including the face of the subject that is captured in an image generated by imaging the subject includes a shielding region that shields at least a part of the region of the face.

[0013] When the shielding region determination unit 11 determines that the face region does not include a shielding region, the first determination unit 12 determines whether the subject is a living body by a first determination method using the image. The first determination method may be a method suitable for determining whether the subject is a living body when the face region does not include a shielding region.

[0014] When the shielding region determination unit 11 determines that the face region includes a shielding region, the second determination unit 13 determines whether the subject is a living body by a second determination method using the image and different from the first determination method. The second determination method may be a method suitable for determining whether the subject is a living body when the face region includes a shielding region.

[0015] The shielding region that shields at least a part of the face in the face region may not be usable for biometric authentication. Therefore, when the face region includes the shielding region, the amount of information that can be obtained from the face region may be less than when the shielding region is not included. Thus, the biometric authentication device 1 in the first embodiment determines whether the subject is a living body by different determination methods depending on whether the face region includes the shielding region or not. [1-2: Technical effects of the biometric authentication device 1]

[0016] Since the biometric authentication device 1 in the first embodiment determines whether the subject is a living body by different determination methods depending on whether the face region includes the shielding region or not, biometric authentication can be performed with high accuracy both when the face region does not include the shielding region and when the face region includes the shielding region. [2: Second embodiment]

[0017] Subsequently, a second embodiment of the biometric authentication device, the biometric authentication method, and the recording medium will be described. Hereinafter, the second embodiment of the biometric authentication device, the biometric authentication method, and the recording medium will be described using a biometric authentication system 2 including a biometric authentication device 20 to which the second embodiment of the biometric authentication device, the biometric authentication method, and the recording medium is applied. [2-1: Configuration of the biometric authentication system 2]

[0018] With reference to FIG. 2, the configuration of the biometric authentication system 2 in the second embodiment will be described. FIG. 2 is a block diagram showing the configuration of the biometric authentication system 2 in the second embodiment. The biometric authentication system 2 in the second embodiment may be applied, for example, to biometric authentication at the time of gate authentication, biometric authentication at the time of walk-through authentication, and the like.

[0019] As shown in FIG. 2, the biological determination system 2 includes a biological determination device 20 and an image generation device 100. The biological determination system 2 may include a single image generation device 100 or may include a plurality of image generation devices 100. The biological determination device 20 and the image generation device 100 can communicate with each other via a communication network. The communication network may include a wired communication network. The communication network may include a wireless communication network. The image generation device 100 is an imaging device capable of imaging a subject located within the imaging target range of the image generation device 100. The subject is a person and is a living body. By imaging the subject, the image generation device 100 generates an image IMG in which the subject imaged by the image generation device 100 is captured. In particular, the image generation device 100 generates an image IMG in which the face of the subject is captured by imaging the face of the subject. The image generation device 100 outputs the generated image IMG to the biological determination device 20. Specifically, the image generation device 100 transmits the generated image IMG to the biological determination device 20 via the communication network.

[0020] The image generation device 100 may generate an image IMG including information indicating the surface body temperature distribution of the subject imaged by the image generation device 100 by imaging the subject. In this case, the image IMG may be an image indicating the surface body temperature distribution of the subject by color or gradation. Alternatively, the image generation device 100 may generate an image IMG including information indicating the distance from the image generation device 100 to the subject by imaging the subject. In this case, the image IMG may be an image indicating the distance information distribution from the image generation device 100 to the subject by color or gradation. The image IMG may be an image such as an infrared image, a thermal image, or a depth image. In the second embodiment, the case where the image generation device 100 generates an image IMG including information indicating the surface body temperature distribution of the subject will be described as an example.

[0021] Furthermore, the image IMG may include an image generated by the image generation device 100 capturing a subject who does not have the intention for the image generation device 100 to capture the subject. The image IMG may also include an image generated by the image generation device 100 capturing a subject who has the intention for the image generation device 100 to capture the subject.

[0022] Based on the received image IMG, the biometric determination device 20 determines whether the subject in front of the image generation device 100 is a living body or a fake of another person pretending to be the subject. Thereby, the biometric determination device 20 can determine whether the subject is pretending to be another person different from the subject. The fake of another person may include a photograph in which another person is captured, a planar object such as a display on which another person is displayed, and a three-dimensional object such as a doll imitating another person, a face mask such as silicon.

[0023] As shown in FIG. 2, the biometric determination device 20 includes an arithmetic device 21 and a storage device 22. Furthermore, the biometric determination device 20 may include a communication device 23, an input device 24, and an output device 25. However, the biometric determination device 20 may not include at least one of the communication device 23, the input device 24, and the output device 25. The arithmetic device 21, the storage device 22, the communication device 23, the input device 24, and the output device 25 may be connected via a data bus 26.

[0024] The arithmetic unit 21 includes, for example, at least one of a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), and an FPGA (Field Programmable Gate Array). The arithmetic unit 21 reads a computer program. For example, the arithmetic unit 21 may read a computer program stored in the storage device 22. For example, the arithmetic unit 21 may read a computer program stored in a computer-readable and non-transitory recording medium using a recording medium reading device (for example, the input device 24 described later) provided in the biological determination device 20. The arithmetic unit 21 may obtain a computer program from a device (not shown) disposed outside the biological determination device 20 via the communication device 23 (or other communication devices) (that is, it may be downloaded or read). The arithmetic unit 21 executes the read computer program. As a result, logical functional blocks for executing the operations to be performed by the biological determination device 20 are realized within the arithmetic unit 21. That is, the arithmetic unit 21 can function as a controller for realizing logical functional blocks for executing the operations (in other words, processes) to be performed by the biological determination device 20.

[0025] FIG. 2 shows an example of logical functional blocks realized in the arithmetic unit 21 for executing the biological determination operation. As shown in FIG. 2, within the arithmetic unit 21, a mask area determination unit 211, which is a specific example of the "shielded area determination means", a first determination unit 212, which is a specific example of the "first determination means", a second determination unit 213, which is a specific example of the "second determination means", an image acquisition unit 214, a face detection unit 215, and a face information detection unit 216 are realized.

[0026] Details of the operations of the mask area determination unit 211, the first determination unit 212, the second determination unit 213, the image acquisition unit 214, the face detection unit 215, and the face information detection unit 216 will be described later with reference to FIGS. 3 to 6.

[0027] The storage device 22 can store desired data. For example, the storage device 22 may temporarily store a computer program executed by the arithmetic device 21. The storage device 22 may temporarily store data that the arithmetic device 21 temporarily uses when the arithmetic device 21 is executing a computer program. The storage device 22 may store data that the biometric determination device 20 stores in the long term. Note that the storage device 22 may include at least one of a RAM (Random Access Memory), a ROM (Read Only Memory), a hard disk device, a magneto-optical disk device, an SSD (Solid State Drive), and a disk array device. That is, the storage device 22 may include a non-temporary recording medium.

[0028] The storage device 22 may store a first determiner 2121 used by the first determination unit 212 for the area determination operation, a first determiner 2122 used by the first determination unit 212 for the spoofing determination operation, a second determiner 2131 used by the second determination unit 213 for the area determination operation, and a second determiner 2132 used by the second determination unit 213 for the spoofing determination operation.

[0029] The first determiner 2121 may be a determiner that determines a first image portion R1 including at least a part of the face region FR in the image IMG. The second determiner 2131 may be a determiner that determines a second image portion R2 from which at least a part of the mask region MR is removed from the first image portion R1. The second determiner 2131 may be a determiner that determines the size and position of the second image portion R2 to be extracted based on the distance between both eyes included in the face. Further, the second determiner 2131 may be a determiner that determines the lower limit portion of the second image portion R2 to be extracted based on both side portions of the lower part of the nasal dorsum included in the face. The first discriminator 2122 may be a discriminator that discriminates whether a subject not wearing a mask is a living body. The first discriminator 2122 may be a discriminator constructed by machine learning for discriminating whether a subject is a living body using a face image of a face not wearing a mask. Further, the second discriminator 2132 may be a discriminator that discriminates whether a subject wearing a mask is a living body. The second discriminator 2132 may be a discriminator constructed by machine learning for discriminating whether a subject is a living body using a face image of a face wearing a mask. That is, two discriminators, namely, the first discriminator 2122 that performs machine learning using a face image of a face not wearing a mask and the second discriminator 2132 that performs machine learning using a face image of a face wearing a mask, may be provided.

[0030] The communication device 23 can communicate with a device outside the biometric authentication device 20 via a communication network (not shown). The communication device 23 may acquire the image IMG used for the biometric authentication operation via the communication network.

[0031] The input device 24 is a device that receives input of information to the biometric authentication device 20 from the outside of the biometric authentication device 20. For example, the input device 24 may include an operating device (for example, at least one of a keyboard, a mouse, and a touch panel) that can be operated by an operator of the biometric authentication device 20. For example, the input device 24 may include a reading device that can read information recorded as data on an externally attachable recording medium with respect to the biometric authentication device 20.

[0032] The output device 25 is a device that outputs information to the outside of the biological determination device 20. For example, the output device 25 may output the information as an image. That is, the output device 25 may include a display device (so-called, display) capable of displaying an image indicating the information to be output. For example, the output device 25 may output the information as voice. That is, the output device 25 may include a voice device (so-called, speaker) capable of outputting voice. For example, the output device 25 may output the information on paper. That is, the output device 25 may include a printing device (so-called, printer) capable of printing desired information on paper. [2-2: Biological determination operation performed by biological determination device 20]

[0033] Subsequently, with reference to FIGS. 3 to 6, the biological determination operation performed by the biological determination device 20 in the second embodiment will be described.

[0034] With reference to FIG. 3, the flow of the biological determination operation performed by the biological determination device 20 in the second embodiment will be described. FIG. 3 is a flowchart showing the flow of the biological determination operation performed by the biological determination device 20 in the second embodiment. Further, FIG. 4 is a conceptual diagram of face information detection in the second embodiment, and FIG. 5 is a conceptual diagram of an image portion in the second embodiment.

[0035] As shown in FIG. 3, the image acquisition unit 214 acquires an image IMG from the image generation device 100 via the communication device 23 and the communication network (step S20).

[0036] The face detection unit 215 detects a face region FR including the face of the subject captured in the image IMG generated by imaging the subject (step S21). The face detection unit 215 may detect, for example, a face region FR including the face of the subject illustrated in FIG. 4(a).

[0037] The face information detection unit 216 detects feature points from a face region FR including the face of the subject (step S22). The face information detection unit 216 may detect feature points indicated by black circles exemplified in FIGS. 4(b) and 4(c) from the face region FR. The black circles exemplified in FIGS. 4(b) and 4(c) may be face feature points. The face feature points may be at least any of points corresponding to the pupil center, the wing of the nose, the edge of the mouth, the contour of the face, the edge of the hair, etc.

[0038] The mask region determination unit 211 determines whether the face region FR includes a shielding region that shields at least a partial region of the face (step S23). The shielding region may be a mask region MR shielded by a mask worn by the face. The mask region determination unit 211 may determine whether the face region FR includes a mask region MR shielded by a mask worn by the face. For example, as exemplified in FIG. 4(b), when feature points such as the wing of the nose and the edge of the mouth are detected from the face region FR, the mask region determination unit 211 may determine that it does not include the mask region MR. Further, for example, as exemplified in FIG. 4(c), when feature points such as the wing of the nose and the edge of the mouth are not detected from the face region FR, the mask region determination unit 211 may determine that it includes the mask region MR. In the second embodiment, the mask region MR hidden by the mask may be a predetermined region including the wing of the nose, the edge of the mouth, etc., and the biological authentication device 20 may perform a biological authentication operation on the premise that the area around the eyes is not hidden by the mask. The first determination device 2122 may determine whether the subject is a living body based on the temperature distribution in the first image portion R1.

[0039] The biological authentication device 20 in the second embodiment may detect whether the subject is wearing a mask and switch the determination operation of the image portion and the spoofing determination operation. That is, the biological authentication device 20 may be provided with a mechanism that appropriately switches the operations by the first determination device 2121 and the first determination device 2122, or the operations by the second determination device 2131 and the second determination device 2132 according to whether the subject is wearing a mask.

[0040] When the mask area determination unit 211 determines that the face area FR does not include a shielding area (step S23: No), the first determination unit 212 determines an area using the first decision maker 2121 (step S24), and uses the first discriminator 2122 to determine whether the subject is a living body or not by a first determination method using the image IMG (step S25). The first determination method may be a method of determining whether the subject is a living body or not using at least a part of the face area FR in the image IMG as a first image portion. The first determination method may be a method of determining whether the subject is a living body or not using a first decision maker 2121 capable of outputting a determination result as to whether the subject is a living body or not when the image IMG is input. The first decision maker 2121 may determine a first image portion R1 to be input to the first discriminator 2122. The first determination unit 212 may determine, for example, an area surrounded by a broken line including an area including feature points, as illustrated in FIG. 5(a), as the first image portion R1.

[0041] When the mask area determination unit 211 determines that the face area FR includes a shielding area (step S23: Yes), the second determination unit 213 determines a second image portion R2 using the second decision maker 2131 (step S26), and uses the second discriminator 2132 to determine whether the subject is a living body or not by a second determination method that uses the image IMG and is different from the first determination method (step S27). The second decision maker 2131 may determine, for example, an area surrounded by a broken line including an area including feature points, as illustrated in FIG. 5(b), as the second image portion R2.

[0042] For example, the second decision maker 2131 may determine, as the second image portion R2, an area above the area determined as the nose among the portions corresponding to the skin in the image IMG. The second decision maker 2131 may use the face information to determine an area below the hair and above the nose, and use it as the second image portion R2 to be input to the second discriminator 2132. [2-3: Determination operation of the second image portion R2 performed by the second decision maker 2131]

[0043] The second determiner 2131 may extract a second image portion R2 from the face region FR based on the distance between the two eyes included in the face. Further, the second determiner 2131 may extract the second image portion R2 from the face region FR based on both sides of the lower part of the nasal dorsum included in the face. A specific method for determining the second image portion R2 when the subject is wearing a mask will be exemplified with reference to FIG. 6. FIG. 6 illustrates a method for determining the second image portion R2 using facial feature points related to the eyes and the nose.

[0044] As shown in FIG. 6(a), first, the coordinates (u 1 coordinates and u 2 coordinates) of the center of the eyes, and the coordinates (v 1 coordinates and v 2 coordinates) of the nose feature points are determined. As the nose feature points, both sides of the lower part of the nasal dorsum, which is the lower limit of the region not shielded by the mask, may be detected.

[0045] As shown in FIG. 6(b), the center (u coordinate) of the fixation region is calculated from the coordinates (u 1 coordinates and u 2 coordinates) of the center of the eyes. As shown in FIG. 6(c), the lower limit (v coordinate) of the fixation region is calculated from the coordinates (v 1 coordinates and v 2 coordinates) of the nose feature points. As shown in FIG. 6(d), the interocular distance (d) is calculated. Using the calculated center (u coordinate) of the fixation region, the lower limit (v coordinate) of the fixation region, and the interocular distance (d), a region where the vertical distance is "d×r" and the horizontal distance is "2×d×r" is set as the second image portion R2. "r" is a parameter for controlling the size of the region.

[0046] The second determination method may be a method for determining whether the subject is a living body using the second image portion R2 from which at least a part of the shielding region has been removed from the first image portion R1. The second determination method may be a method for determining whether the subject is a living body using a second determiner 2132 capable of outputting a determination result as to whether the subject is a living body when the image IMG is input. The second determiner 2132 may determine whether the subject is a living body based on the temperature distribution within the second image portion R2.

[0047] Furthermore, as a method for determining the viewing area, in addition to the method using facial feature points, for example, a method using segmentation information may be adopted. The face segmentation information is information indicating the positions of pixels where eyes, nose, etc. exist.

[0048] Also, the first determiner 2121 and the first discriminator 2122 may not be separate mechanisms but may be integrated. Also, the second determiner 2131 and the second discriminator 2132 may not be separate mechanisms but may be integrated. [2-5: Technical Effects of the Biometric Authentication Device 20]

[0049] The biometric authentication device 20 in the second embodiment determines whether the subject is a living body by utilizing the characteristic that the features shown in the image of the living body are different from the features shown in the image of a forgery that is not a living body. However, the features shown in the image in the shielding area that shields at least a part of the face area may not show the features shown in the image of the living body and may be similar to the features shown in the image of a forgery that is not a living body. For this reason, if the image portion of the shielding area is used for biometric authentication, there is a risk of false determination.

[0050] Therefore, when the face area FR includes a shielding area, the biometric authentication device 20 in the second embodiment uses a second image portion R2 obtained by removing at least a part of the shielding area from an image portion including at least a part of the face area FR in the image IMG to determine whether the subject is a living body. Thereby, even when the face area FR includes a shielding area, it is possible to accurately determine whether it is a living body.

[0051] Recently, there has been a demand for realizing biometric authentication and biometric identification while wearing a mask. The mask area MR shielded by the mask worn on the face corresponds to a shielding area that shields at least a part of the face area. The biometric authentication system 2 in the second embodiment can also realize accurate biometric authentication even when there is a mask area MR shielded by the mask worn on the face.

[0052] When determining whether a subject is a living body using a second image portion R2 from which at least a part of a mask region MR has been removed from an image IMG, the second image portion R2 may be extracted from the image IMG based on the distance between both eyes included in the face. Thereby, it is possible to determine whether the subject is a living body using an image portion appropriate for living body determination. (Features shown in the image of the living body)

[0053] On the other hand, when performing a living body determination using an image portion around the eyes that is not blocked by the mask when the face region FR does not include the mask region MR, the amount of available information decreases compared to the living body determination using the face region FR as the image portion. Therefore, the accuracy of the living body determination may deteriorate. The living body determination device 20 in the second embodiment determines whether the subject is a living body using a first image portion R1 that includes at least a part of the face region FR in the image IMG when the face region FR does not include the mask region MR. Thus, it is possible to accurately determine whether it is a living body.

[0054] Also, it is conceivable that preferable living body determination operations differ depending on the image used. Therefore , the living body determination device 20 in the second embodiment prepares a first determination device to be used when there is no shielding region that shields the face, and a second determination device to be used when there is a shielding region that shields the face. By switching the determination device to be used according to whether there is a shielding region and making the living body determination operation different, it is possible to accurately perform the living body determination.

[0055] As described above, the living body determination device 20 in the second embodiment can accurately perform the living body determination and can realize highly accurate spoof detection. Therefore, the safety of the face authentication technology can be ensured. [3: Third Embodiment]

[0056] Next, a third embodiment of the biological authentication device, the biological authentication method, and the recording medium will be described. Hereinafter, the third embodiment of the biological authentication device, the biological authentication method, and the recording medium will be described using a biological authentication system 3 including a biological authentication device 30 to which the third embodiment of the biological authentication device, the biological authentication method, and the recording medium is applied. [3-1: Configuration of the biological authentication system 3]

[0057] The biological authentication system 3 in the third embodiment may have the same configuration as the biological authentication system 2 in the above-described second embodiment. The biological authentication system 3 in the third embodiment uses a different face image for the biological authentication operation compared to the biological authentication system 2 in the second embodiment. Specifically, the method for acquiring the face image used by the biological authentication system 3 in the third embodiment is different. More specifically, the lighting control by the image generation device 100 is different. The image generation device 100 in the third embodiment may be a device including a lighting device and a camera. [3-2: Light emission operation performed by the lighting device]

[0058] The lighting device can illuminate the subject with illumination light. The lighting device can change the lighting conditions. The lighting conditions may include, for example, conditions regarding the presence or absence of light emission of the illumination light. The lighting conditions may include the intensity of the illumination light. The lighting conditions may include the illumination direction of the illumination light (that is, the direction in which the illumination light is irradiated onto the subject, and the direction of the light source of the illumination light as seen from the subject).

[0059] For example, a display emits light for the purpose of displaying an image (note that the image may mean at least one of a still image and a moving image), and the light emitted by this display may be used as illumination light for illuminating the subject. In this case, the display D of a smartphone or tablet terminal held by the subject may be used as the lighting device. The image generation device 100 may be a smartphone or tablet terminal held by the subject.

[0060] When the display D is used as a lighting device, the lighting conditions (particularly, the conditions regarding the presence or absence of emission of illumination light) may include the conditions regarding the on / off of the display D. The lighting conditions (particularly, the conditions regarding the intensity of illumination light) may include the conditions regarding the brightness (i.e., luminance) of the display surface of the display D. The lighting conditions (particularly, the conditions regarding the intensity of illumination light) may include the conditions regarding the change in the brightness (i.e., luminance) of the display surface of the display D.

[0061] The lighting conditions may include the conditions regarding the display mode of each display area when the display surface of the display D is divided into a plurality of display areas. The display mode of the display area may include the brightness (i.e., luminance) of the display area. The display mode of the display area may include the change in the brightness (i.e., luminance) of the display area. Note that the lighting conditions may be set so that the display modes of all the plurality of display areas are the same. Alternatively, the lighting conditions may be set so that the display modes of at least two of the plurality of display areas are different from each other.

[0062] The display D used as a lighting device may change the lighting conditions between the OFF condition where the display D does not emit light (i.e., the display is in the off state) and the ON condition where the display D emits light (i.e., the display is in the on state). The display D used as a lighting device may change the lighting conditions between the condition where the color of the display surface of the display D becomes the first color (i.e., light having a wavelength presenting the first color is emitted from the display surface of the display D) and the condition where the color of the display surface of the display D becomes the second color different from the first color (i.e., light having a wavelength presenting the second color is emitted from the display surface of the display D). Further, the color of the display D may be changed, for example, from orange to blue to detect the change in the color of a specific face area such as the forehead or cheeks, and according to the detection result, the uneven state of the face area, materials such as skin or silicon, etc. may be detected, and a liveness determination may be performed. The change in color may be at least one change in R, G, or B.

[0063] Further, when conditions regarding the display mode of each display area in the case where the display surface of the display D is divided into a plurality of display areas are used as illumination conditions, the display D used as the illumination device may change the illumination conditions between a first display condition in which the combination of the display modes of the plurality of display areas is set to a first combination and a second display condition in which the combination of the display modes of the plurality of display areas is set to a second combination different from the first combination. For example, the display D used as the illumination device may change the illumination conditions between a first display condition in which the luminance of a part of the plurality of display areas (for example, the right half of the display surface of the display D) is set to a first luminance and the luminance of another part of the plurality of display areas (for example, the left half of the display surface of the display D) is set to a second luminance darker than the first luminance, and a second display condition in which the luminance of a part of the plurality of display areas (for example, the right half of the display surface of the display D) is set to the second luminance and the luminance of another part of the plurality of display areas (for example, the left half of the display surface of the display D) is set to the first luminance.

[0064] As a specific example, FIGS. 7(a) to 7(d) show the illumination conditions in the case where the display surface of the display D is divided into two display areas L and R arranged in the horizontal direction (left - right direction) of the display surface. FIG. 7(a) shows an illumination condition in which both of the two display areas L and R emit light (or the luminance of both of the two display areas L and R is set to a relatively bright luminance). FIG. 7(b) shows an illumination condition in which neither of the two display areas L and R emits light (or the luminance of both of the two display areas L and R is set to a relatively dark luminance). FIG. 7(c) shows an illumination condition in which the display area L emits light (or the luminance of the display area L is set to a relatively bright luminance), while the display area R does not emit light (or the luminance of the display area R is set to a relatively dark luminance). FIG. 7(d) shows an illumination condition in which the display area R emits light (or the luminance of the display area R is set to a relatively bright luminance), while the display area L does not emit light (or the luminance of the display area L is set to a relatively dark second luminance). [3 - 3: Determination operation of the image part]

[0065] The image acquisition unit 214 acquires a plurality of person images respectively generated by imaging a subject under a plurality of different illumination conditions. The mask area determination unit 211 determines whether or not a face area including the face of the subject imaged in at least one of the plurality of person images includes a shielding area that shields at least a partial area of the face. The operation of determining an image portion by the biological authentication device 30 in the third embodiment may be the same as the operation of determining an image portion by the biological authentication device 20 in the second embodiment shown in FIG. 3. [3-4: Biological authentication operation performed by the biological authentication device 30]

[0066] In the third embodiment, the biological authentication device 30 can determine whether or not a subject is a living body by having the subject image a forged object of another person that the subject is impersonating with a camera. In this case, the biological authentication device 30 may determine whether or not the subject is a living body by utilizing the characteristics of diffuse reflection in which the characteristics of reflected light from a three-dimensional living body (for example, reflected light of at least one of illumination light and ambient light, the same applies hereinafter) are different from the characteristics of reflected light from a forged object that is not a living body.

[0067] For example, the biological authentication device 30 may calculate a feature amount that reflects the three-dimensional shape of the face of the subject imaged in the image IMG based on the image IMG. When the subject is a living body, since the camera images the subject, the living body is imaged in the image IMG. Therefore, in this case, the biological authentication device 30 calculates a feature amount corresponding to the three-dimensional shape (for example, a feature amount representing three-dimensional features) from the image IMG in which the living body is imaged. On the other hand, when the subject is impersonating another person with a forged object such as a photograph or a display, since the camera images the forged object, the forged object is imaged in the image IMG. Therefore, in this case, the biological authentication device 30 calculates a feature amount corresponding to, for example, a planar shape (for example, a feature amount representing planar features) from the image IMG in which the forged object is imaged. Thus, the biological authentication device 30 may determine whether or not the subject is impersonating another person based on the calculated feature amount.

[0068] The living body determination system 3 may include the living body determination device 30 and the image generation device 100 as separate devices. Alternatively, the living body determination system 3 may be provided as a single device including the living body determination device 30 and the image generation device 100. For example, the living body determination system 3 may include a smartphone or a tablet terminal including the living body determination device 30 and the image generation device 100. [3-5: Technical effects of the living body determination device 30]

[0069] In the third embodiment, the living body determination device 30 determines whether the subject is a living body by using a plurality of person images IMG respectively generated by imaging the subject under a plurality of different illumination conditions. That is, the living body determination device 30 in the third embodiment determines whether the subject is a living body by utilizing the diffuse reflection characteristic that the characteristics of the reflected light from a three-dimensional living body are different from the characteristics of the reflected light from a non-living fake object. On the other hand, the characteristics of the reflected light from the mask region MR may be more similar to the characteristics of the reflected light from a non-living fake object than the characteristics of the reflected light from a living body. For this reason, it is not preferable to use the image portion based on the reflected light from the mask region MR for living body determination. Even when the living body determination device 30 in the third embodiment uses a plurality of person images IMG respectively generated by imaging the subject under a plurality of different illumination conditions, the determination operation is changed according to whether the face region FR includes the mask region MR, so that it is possible to accurately determine whether the subject is a living body. [4: Fourth Embodiment]

[0070] Subsequently, a fourth embodiment of the living body determination device, the living body determination method, and the recording medium will be described. Hereinafter, the fourth embodiment of the living body determination device, the living body determination method, and the recording medium will be described using an authentication system 4 including an authentication device 40 to which the fourth embodiment of the living body determination device, the living body determination method, and the recording medium is applied.

[0071] The authentication system 4 may be used, for example, to manage the entry and exit of a target person to / from a restricted area. Specifically, the restricted area is an area where entry of a target person who satisfies a predetermined entry condition is permitted, while entry of a target person who does not satisfy the predetermined entry condition is not permitted (i.e., prohibited). In this case, the authentication device 40 may authenticate the target person by determining whether the target person captured in the image IMG is the same as a person permitted to enter the restricted area (for example, a person registered in advance as a person who satisfies the entry condition). In this case, when it is determined that the target person is the same as a person permitted to enter the restricted area (i.e., when the authentication is successful), the authentication device 40 may permit the target person to enter the restricted area. As an example, the authentication device 40 may set the state of an entry / exit restriction device (for example, a gate device or a door device) capable of restricting the passage of the target person to an open state in which the target person can pass through the entry / exit restriction device. On the other hand, when it is determined that the target person is not the same as a person permitted to enter the restricted area (i.e., when the authentication fails), the authentication device 40 may prohibit the target person from entering the restricted area. As an example, the authentication device 40 may set the state of the entry / exit restriction device to a closed state in which the target person cannot pass through the entry / exit restriction device. Further, even when the authentication of the target person is successful, if it is determined that the target person captured in the image IMG is not a living body, the authentication system 4 may prohibit the target person from entering the restricted area.

[0072] When the authentication system 4 is used to manage the entry and exit of a target person to and from a restricted area, the image generation device 100 may image the target person who is attempting to enter the restricted area. As an example, the image generation device 100 may be arranged near the entry / exit restriction device and image the target person who is located near the entry / exit restriction device in order to enter the restricted area. In this case, the image generation device 100 may image the target person who is moving towards the entry / exit restriction device. The image generation device 100 may image the target person who is moving towards the image generation device 100 arranged near the entry / exit restriction device. Alternatively, the image generation device 100 may image the target person who is stationary in front of the entry / exit restriction device. The image generation device 100 may image the target person who is stationary in front of the image generation device 100 arranged near the entry / exit restriction device. [4-1: Configuration of Authentication Device 40]

[0073] As shown in FIG. 8, the authentication device 40 in the fourth embodiment is different from the biometric determination device 20 in the second embodiment (or compared with the biometric determination device 30 in the third embodiment) in that the arithmetic device 21 includes an authentication unit 417, and the storage device 22 stores a third decision maker 4171, a first authenticator 4172, and a second authenticator 4173. The authentication unit 417 authenticates a subject based on an image. The third decision maker 4171 may be a decision maker that determines a third image portion R3 used by the authentication unit 417 to authenticate the subject among the images IMG. The first authenticator 4172 may be an authenticator that authenticates a subject not wearing a mask. The first authenticator 4172 may be an authenticator constructed by machine learning for authenticating the subject using a face image of a face without a mask. Further, the second authenticator 4173 may be an authenticator that authenticates a subject wearing a mask. The second authenticator 4173 may be an authenticator constructed by machine learning for authenticating the subject using a face image of a face wearing a mask. That is, two authenticators, namely, the first authenticator 4172 that performs machine learning using a face image of a face without a mask and the second authenticator 4173 that performs machine learning using a face image of a face wearing a mask, may be provided. Other features of the authentication device 40 may be the same as other features of the biometric determination device 20 (or the biometric determination device 30).

[0074] For example, when the first determination unit 212 or the second determination unit 213 performs biometric determination using an image captured by a visible light camera, the image generation device 100 may include a visible light camera. In this case, the first determination unit 212 or the second determination unit 213 may perform biometric determination based on an image captured by the visible light camera, and the authentication unit 417 may also authenticate the subject based on an image captured by the visible light camera. Also, the biometric determination using an image captured by the visible light camera may be referred to as luminance difference liveness.

[0075] When one of them, for example, the first determination unit 212 or the second determination unit 213, performs biometric determination using an infrared image, the image generation device 100 may include an infrared camera and a visible light camera. In this case, the first determination unit 212 or the second determination unit 213 may perform biometric determination based on the infrared image captured by the infrared camera, and the authentication unit 417 may authenticate the subject based on the image captured by the visible light camera. Also, the biometric determination using the infrared image may be referred to as infrared image liveness. [4-2: Biometric determination operation and biometric authentication operation performed by the authentication device 40]

[0076] Hereinafter, with reference to FIG. 9, the flow of the biometric determination operation and the biometric authentication operation in the fourth embodiment will be described. FIG. 9 is a flowchart showing the flow of the biometric determination operation and the biometric authentication operation in the fourth embodiment. Hereinafter, the case where the first determination unit 212 or the second determination unit 213 performs luminance difference liveness will be described as an example.

[0077] The operations in steps S20 to S27 may be the same as the operations in steps S20 to S27 of the second embodiment or the third embodiment shown in FIG. 3. That is, also in the fourth embodiment, spoofing determination described in at least one of the second embodiment and the third embodiment may be performed.

[0078] When the mask area determination unit 211 determines that the face area FR does not include a shielding area (step S23: No), in parallel with the operations in steps S24 and S25, the operations in steps S41 and S42 are executed. That is, the authentication unit 417 determines the third image portion R3 using the third decision maker 4171 (step S41), and authenticates the subject by the first authentication method using the first authenticator 4172 (step S42).

[0079] Also, when the mask area determination unit 211 determines that the face area FR includes a shielding area (step S23: Yes), the operations of steps S43 and S44 are executed in parallel with the operations of steps S26 and S27. That is, the authentication unit 417 determines the third image portion R3 using the third determiner 4171 (step S43), and authenticates the subject by the second authentication method using the second authenticator 4173 (step S44). In the fourth embodiment, the third image portion R3 used for authentication may be the same regardless of whether a mask is worn or not.

[0080] If it is determined that the subject is a living body in the spoofing determination in step S25 and the authentication operation in step S42 is successful, or if it is determined that the subject is a living body in the spoofing determination in step S27 and the authentication operation in step S44 is successful, it is determined that the authentication of the subject is successful (step S45). That is, if either one of the operations fails, it is determined that the authentication of the subject has failed.

[0081] In the above embodiment, the case where the spoofing determination operation and the authentication operation are executed in parallel has been described. However, for example, the operations may be performed in the order of step S24, step S25, step S41, step S42, or in the order of step S26, step S27, step S43, step S44.

[0082] Also, for example, before the determination in step S23, the first authentication or the second authentication by the authentication unit 417 may be performed. In this case, when the first authentication or the second authentication by the authentication unit 417 is completed, the determination in step 23 and the operations of steps S24 and S25, or the operations of steps S26 and S27 may be performed. Alternatively, for example, when the first authentication or the second authentication by the authentication unit 417 is completed, the operations of steps S24 and S25, or the operations of steps S26 and S27 may be performed. Since the process proceeds to other operations only when the first authentication or the second authentication by the authentication unit 417 is completed, unnecessary processing can be reduced.

[0083] Further, for example, before the determination in step S23, biometric determination by the first determination unit 212 or the second determination unit 213 may be performed. In this case, when the biometric determination by the first determination unit 212 or the second determination unit 213 is completed, the determination in step 23 and the operations in steps S41 and S42, or the operations in steps S43 and S44 may be performed. Or, for example, when the biometric determination by the first determination unit 212 or the second determination unit 213 is completed, the operations in steps S41 and S42, or the operations in steps S43 and S44 may be performed. Since the transition to other operations is made only when the biometric determination by the first determination unit 212 or the second determination unit 213 is completed, wasteful processing can be reduced.

[0084] Note that the mask area determination by the mask area determination unit 211 and the face authentication by the authentication unit 417 may be performed based on the same one image. In this case, the mask area determination operation and the face authentication operation may be performed prior to the biometric determination operation. The biometric determination operation by the first determination unit 212 or the second determination unit 213 may start when a predetermined number of images are acquired.

[0085] Also, when the first determination unit 212 or the second determination unit 213 performs infrared image liveness, since the camera that captures the image used for the liveness determination operation is different from the camera that captures the image used for face authentication, the timing at which the liveness determination operation and the face authentication operation start, and the timing at which each operation is completed may be different. [4-3: Conceptual diagram of the image part]

[0086] In the fourth embodiment, the size of the image part used by the first determination unit 212 among the images to determine whether the subject is a living body may be larger than the size of the image part used by the authentication unit 417 among the images to authenticate the subject. The first determination unit 212 may extract the first image part R1 from the face region FR so as to be larger than the size of the image part used by the authentication unit 417 to authenticate the subject.

[0087] Figure 10 is a conceptual diagram of an image portion in the fourth embodiment. In Figure 10, the region R1 enclosed by the two-dot chain line indicates a first image portion that includes at least a part of the face region FR in the image IMG, and exemplifies an image portion used for the impersonation determination operation of a subject not wearing a mask. Since a larger amount of information that can be used for biometric authentication can be obtained by extracting a larger portion of the skin in the face region FR that is not shielded, it is preferable that the image portion used for biometric authentication be larger. Also, portions that are easily covered by hair can be considered as shielded regions and may be excluded from the image portion.

[0088] The region R3 enclosed by the broken line indicates a third image portion that the authentication unit 417 in the image IMG uses to authenticate the subject, and exemplifies an image portion used for the authentication operation. That is, the first image portion R1 may include another image portion of the face region FR in addition to the third image portion R3. Thereby, the image portion that can be used for biometric authentication can be enlarged, and the accuracy of biometric authentication can be improved.

[0089] Also, as an example of liveness determination, a method can be cited in which a pulse is estimated based on the difference between images of a plurality of continuously captured images, and it is determined whether the subject is a living body. Even when this method is adopted, the accuracy of biometric authentication can be improved by enlarging the image portion that can be used.

[0090] Also, the region R2 enclosed by the two-dot chain line indicates a second image portion from which at least a part of the mask region MR is removed from the first image portion R1, and exemplifies an image portion used for the impersonation determination operation of a face image wearing a mask. [4-3: Technical Effects of the Authentication Device 40]

[0091] Since the authentication device 40 in the fourth embodiment performs the biometric authentication described in at least one of the second and third embodiments, it is possible to more accurately determine whether the subject is a living body, and it is possible to accurately authenticate a living subject as the person himself / herself. Also, in the authentication device 40 according to the fourth embodiment, the size of the image portion used by the first determination unit 212 in the image to determine whether the subject is a living body is made larger than the size of the image portion used by the authentication unit 417 in the image to authenticate the subject, so that the accuracy of the living body determination can be improved.

[0092] In addition, in each of the above embodiments, the case where the shielding area is the mask area shielded by the mask worn on the face has been described. However, even when there is a shielding area shielded by something other than the mask, each of the above embodiments can be applied.

[0093] Also, in each of the above embodiments, the case where the living body determination device is equipped with one type of living body determination mechanism and performs one type of living body determination has been described. However, the living body determination device may be equipped with a plurality of types of living body determination mechanisms, perform a plurality of types of living body determinations, and determine whether it is a living body based on a plurality of determination results.

[0094] The plurality of types of living body determinations may include at least one of a first living body determination that does not switch at least one of the image portion used for the living body determination and the determination device used for the living body determination according to whether at least a part of the face area is shielded, and a second living body determination that switches at least one of the image portion used for the living body determination and the determination device used for the living body determination. That is, when performing the first living body determination, at least one of the image portion used for the living body determination and the determination device used for the living body determination may not be switched according to whether at least a part of the face area is shielded. On the other hand, when performing the second living body determination, at least one of the image portion used for the living body determination and the determination device used for the living body determination may be switched according to whether at least a part of the face area is shielded.

[0095] That is, the biological determination device may include: a first biological determination means for determining whether the subject is a living body using the first image portion R1 when the face region is shielded; and a second biological determination means for determining whether the subject is a living body using the second image portion R2 when the face region is shielded. In this case, the second biological determination means includes a first determination unit 212 and a second determination unit 213.

[0096] Examples of the first biological determination include methods of instructing the subject to perform an operation such as shaking the head and detecting the subject's operation, methods of detecting changes in the subject's line of sight, methods of detecting changes in the reflection of the iris of the subject's eyes, and the like. Examples of the second biological determination include biological determination based on the temperature distribution within the image portion, luminance difference liveness, infrared image liveness, and the like.

[0097] When performing the second biological determination and at least a part of the face region is not shielded, and when performing the first biological determination, for example, the biological determination may be performed using the first image portion R1 shown in FIG. 10. On the other hand, when performing the second biological determination and at least a part of the face region is shielded, for example, the biological determination may be performed using the second image portion R2 shown in FIG. 10. Also, since the images used by the biological determination device differ depending on the type of biological determination to be performed, the image generation device may be equipped with an infrared camera and a visible light camera. [5: Supplementary Note]

[0098] Regarding the embodiments described above, the following supplementary notes are further disclosed. [Supplementary Note 1] Shielding region determination means for determining whether a face region including the face of the subject captured in the image generated by imaging the subject includes a shielding region that shields at least a part of the face; When the shielding region determination means determines that the face region does not include the shielding region, first determination means for determining whether the subject is a living body by a first determination method using the image; When the shielding area determination means determines that the face area includes the shielding area, a second determination means for determining whether the subject is a living body by using the image and by a second determination method different from the first determination method, and A living body determination device comprising the same. [Appendix 2] The first determination method is a method of determining whether the subject is a living body by using a first image portion including at least a part of the face area in the image, The second determination method is a method of determining whether the subject is a living body by using a second image portion obtained by removing at least a part of the shielding area from the first image portion. The living body determination device according to Appendix 1. [Appendix 3] The second determination means extracts the second image portion from the image based on the distance between both eyes included in the face. The living body determination device according to Appendix 2. [Appendix 4] The living body determination device further includes a determination means for determining the size and position of the second image portion to be extracted based on the distance between both eyes included in the face. The living body determination device according to Appendix 3. [Appendix 5] The second determination means extracts the second image portion from the image based on both sides of the lower part of the nasal dorsum included in the face. The living body determination device according to any one of Appendices 2 to 4. [Appendix 6] The living body determination device further includes a determination means for determining the lower limit of the second image portion to be extracted based on both sides of the lower part of the nasal dorsum included in the face. The living body determination device according to Appendix 5. [Appendix 7] The living body determination device further includes an authentication means for authenticating the subject based on the image, The first determination means extracts the first image portion from the image so as to be larger than the size of the image portion used by the authentication means to authenticate the subject. The living body determination device according to any one of Appendices 2 to 6. [Appendix 8] When the face area is shielded, a first biometric determination means for determining whether the subject is a living body using the first image portion; When the face area is shielded, further comprising a second biometric determination means for determining whether the subject is a living body using the second image portion; The second biometric determination means includes the first determination means and the second determination means The biometric determination device according to any one of Appendices 2 to 7. [Appendix 9] The image is an infrared image The biometric determination device according to any one of Appendices 1 to 8. [Appendix 10] The first determination method is a method for determining whether the subject is a living body using a first determiner capable of outputting a determination result as to whether the subject is a living body when the image is input, The second determination method is a method for determining whether the subject is a living body using a second determiner capable of outputting a determination result as to whether the subject is a living body when the image is input and different from the first determiner. The biometric determination device according to any one of Appendices 1 to 9. [Appendix 11] The shielding area is a mask area shielded by a mask worn on the face The biometric determination device according to any one of Appendices 1 to 10. [Appendix 12] Further comprising an image acquisition means for acquiring a plurality of person images respectively generated by imaging a subject using a plurality of different illumination conditions, The image is at least one person image among the plurality of person images The biometric determination device according to any one of Appendices 1 to 11. [Appendix 13] Further comprising an authentication means for authenticating the subject based on the image, The size of the image portion among the said images used by the said first determination means for determining whether the said subject is a living body is larger than the size of the image portion among the said images used by the said authentication means for authenticating the said subject. The living body determination device according to any one of Appendices 1 to 12. [Appendix 14] Determine whether the face area including the face of the said subject imaged in the image generated by imaging the said subject includes a shielding area that shields at least a part of the area of the said face. When it is determined that the said face area does not include the said shielding area, determine whether the said subject is a living body by the first determination method using the said image. When it is determined that the said face area includes the said shielding area, determine whether the said subject is a living body by a second determination method that uses the said image and is different from the said first determination method. Living body determination method. [Appendix 15] On a computer, Determine whether the face area including the face of the said subject imaged in the image generated by imaging the said subject includes a shielding area that shields at least a part of the area of the said face. When it is determined that the said face area does not include the said shielding area, determine whether the said subject is a living body by the first determination method using the said image. When it is determined that the said face area includes the said shielding area, determine whether the said subject is a living body by a second determination method that uses the said image and is different from the said first determination method. A recording medium on which a computer program for executing a living body determination method is recorded.

[0099] At least some of the constituent elements of each of the above-described embodiments can be appropriately combined with at least some of the other constituent elements of each of the above-described embodiments. Some of the constituent elements of each of the above-described embodiments may not be used. Also, to the extent permitted by law, the disclosures of all the documents (for example, published gazettes) cited in this disclosure above are incorporated by reference to form part of the description of this disclosure.

[0100] This disclosure can be appropriately modified within the scope that does not conflict with the technical idea that can be read from the claims and the entire specification. A biological determination device, a biological determination method, and a recording medium involving such modifications are also included in the technical idea of this disclosure.

Explanation of Signs

[0101] 1 Biological determination device 11 Shielded area determination unit 12 First determination unit 13 Second determination unit 2 Biological determination system 20 Biological determination device 100 Image generation device 211 Mask area determination unit 212 First determination unit 2121 First decision maker 2122 First discriminator 213 Second determination unit 2131 Second decision maker 2132 Second discriminator 214 Image acquisition unit 215 Face detection unit 216 Face information detection unit 4 Authentication system 40 Authentication device 417 Authentication unit 4171 Third decision maker 4172 First authenticator 4173 Second authenticator IMG Image FR Face area MR Mask area R1 First image part R2 Second image part R3 Third image part D Display

Claims

1. Occlusion area determination means for determining whether at least a part of the face area including the face of the subject in the image is occluded; When the face area is not occluded, first determination means for determining whether the subject is a living body by a first determination method using a first discriminator that has been machine-learned using a face image of a face without a mask, which can output a determination result as to whether the subject is a living body when the image is input; When the face area is occluded, second determination means for determining whether the subject is a living body by a second determination method using a second discriminator that has been machine-learned using a face image of a face wearing a mask, which can output a determination result as to whether the subject is a living body when the image is input; Authentication means for authenticating the subject based on the image; Comprising; The first determination means extracts, from the image, a first image portion that includes the image portion and includes the face area in the image so as to be larger than the size of the image portion including both the outer ends of the eyebrows, both the outer ends of the eyes, and the jaw included in the face used by the authentication means to authenticate the subject. Living body determination device.

2. The first determination method is a method of inputting the first image portion into the first discriminator to determine whether the subject is a living body; The second determination method is a method of inputting a second image portion obtained by removing the occluded area from the first image portion into the second discriminator to determine whether the subject is a living body. The living body determination device according to Claim 1.

3. It comprises determination means for determining the size of the second image portion based on the distance between both eyes included in the face. The living body determination device according to Claim 2.

4. It comprises determination means for determining the position of the second image portion so that the central portion between both eyes included in the face becomes the central portion of the second image portion. The living body determination device according to Claim 2 or 3.

5. It comprises determination means for determining both side portions at the lower part of the nasal dorsum included in the face as the lower limit portion of the second image portion. The living body determination device according to Claim 2 or 3.

6. First living body determination means for determining whether the subject is a living body by a third determination method using a third discriminator that can output a determination result as to whether the subject is a living body when the first image portion is input; When the face area is not blocked, it is determined whether the subject is a living body by the first determination method, and when the face area is blocked, it further includes a second living body determination means for determining whether the subject is a living body by the second determination method The living body determination device according to claim 2 or 3.

7. A living body determination method executed by a computer, Determining whether at least a part of the face area including the face of the subject in the image is blocked, When the face area is not blocked, when the image is input, the first determination method using a first determiner that has learned machine learning using a face image of a face without wearing a mask, which can output a determination result as to whether the subject is a living body, is used to determine whether the subject is a living body, When the face area is blocked, when the image is input, the second determination method using a second determiner that has learned machine learning using a face image of a face wearing a mask, which can output a determination result as to whether the subject is a living body, is used to determine whether the subject is a living body, Authenticating the subject based on the image, Including, Determining by the first determination method includes extracting, from the image, a first image portion that includes the image portion and includes the face area of the image so as to be larger than the size of the image portion including both eyebrow ends, both eye ends, and the jaw included in the face used to authenticate the subject, Living body determination method.

8. To the computer, Determining whether at least a part of the face area including the face of the subject in the image is blocked, When the face area is not blocked, when the image is input, the first determination method using a first determiner that has learned machine learning using a face image of a face without wearing a mask, which can output a determination result as to whether the subject is a living body, is used to determine whether the subject is a living body, When the face area is blocked, when the image is input, the second determination method using a second determiner that has learned machine learning using a face image of a face wearing a mask, which can output a determination result as to whether the subject is a living body, is used to determine whether the subject is a living body, Authenticating the subject based on the image, Including, Determining by the first determination method means extracting, from the image, a first image portion that includes the face region of the image and that encompasses the image portion including both the outer ends of the eyebrows, both the outer ends of the eyes, and the chin included in the face used for authenticating the subject person, such that the size of the first image portion is larger than the size of the image portion including both the outer ends of the eyebrows, both the outer ends of the eyes, and the chin included in the face used for authenticating the subject person. A computer program for causing a living body determination method to be executed.

Images