Impersonation detection system, impersonation detection method and program

Abstract

To detect impersonation without causing increased costs and system complication.SOLUTION: An impersonation detection system according to an embodiment detects impersonation to another person by a person passing through a predetermined passage. The impersonation detection system comprises an acquisition unit, a face detection unit, and a determination unit. The acquisition unit acquires a first captured image taken from diagonally in front of a person entering the passage. The face detection unit detects a face present in the first captured image. The determination unit determines that impersonation is being performed when two or more faces are detected in one first captured image, and a distance between a first position corresponding to one face of the two faces in the first captured image and a second position corresponding to the other face of the two faces is less than a predetermined threshold value.SELECTED DRAWING: Figure 5

Description

[Technical field]

[0001] An embodiment of the present invention relates to a spoofing detection system, a spoofing detection method, and a program. [Background technology]

[0002] In face authentication systems that detect faces from images (including videos) captured by an imaging device and perform identity verification, etc. based on the detection results, technological development is underway to detect impersonation using photographs of other people's faces, etc. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] International Publication No. 2020 / 152917 [Patent Document 2] Patent No. 5445460 [Patent Document 3] JP 2007-241402 A Summary of the Invention [Problem to be solved by the invention]

[0004] In conventional technology, multiple images captured from different imaging directions and distance information are required, which necessitates multiple cameras and distance sensors, resulting in problems such as increased costs and complicated systems.

[0005] Therefore, an object of the embodiments of the present invention is to provide an spoofing detection system, an spoofing detection method, and a program that are capable of detecting spoofing with high accuracy without increasing costs or complicating the system. [Means for solving the problem]

[0006] An spoofing detection system according to an embodiment detects spoofing of another person by a person passing through a predetermined passage. The spoofing detection system includes an acquisition unit, a face detection unit, and a determination unit. The acquisition unit acquires a first captured image in which a person who has entered the passage is captured from an oblique front. The face detection unit detects a face present in the first captured image. The determination unit determines that spoofing is occurring when two or more faces are detected in one first captured image, and a distance between a first position corresponding to one of the two faces in the first captured image and a second position corresponding to the other of the two faces is smaller than a predetermined threshold. [Brief description of the drawings]

[0007] [Figure 1] FIG. 1 is a diagram illustrating an example of the configuration of a spoofing detection system according to the first embodiment. [Diagram 2] FIG. 2 is a side view showing an example of a situation in which the spoofing detection system of the first embodiment is applied. [Diagram 3] FIG. 3 is a front view showing an example of a situation in which the spoofing detection system of the first embodiment is applied. [Figure 4] FIG. 4 is a top view showing an example of a situation in which the spoofing detection system of the first embodiment is applied. [Diagram 5] FIG. 5 is a diagram illustrating an example of a functional configuration of the spoofing detection system according to the first embodiment. [Figure 6] FIG. 6 is a diagram showing an example of the first captured image according to the first embodiment. [Figure 7] FIG. 7 is a diagram illustrating an example of a method for calculating the coordinates of the first position and the second position in the first embodiment. [Figure 8] FIG. 8 is a diagram showing an example of a situation when a preparatory image used when setting a threshold value in the first embodiment is acquired. [Figure 9] FIG. 9 is a flowchart illustrating an example of a process for setting a threshold value according to the first embodiment. [Figure 10]FIG. 10 is a flowchart showing an example of processing in the spoofing detection system of the first embodiment. [Figure 11] FIG. 11 is a diagram illustrating an example of a functional configuration of the spoofing detection system according to the second embodiment. [Figure 12] FIG. 12 is a top view showing an example of a situation in which the spoofing detection system of the second embodiment is applied. [Figure 13] FIG. 13 is a flowchart showing an example of processing in the spoofing detection system of the second embodiment. [Figure 14] FIG. 14 is a diagram illustrating an example of a functional configuration of the spoofing detection system according to the third embodiment. [Figure 15] FIG. 15 is a flowchart showing an example of processing in the spoofing detection system of the third embodiment. [Figure 16] FIG. 16 is a diagram illustrating an example of a functional configuration of the spoofing detection system according to the fourth embodiment. As illustrated in FIG. [Figure 17] FIG. 17 is a flowchart showing an example of processing in the spoofing detection system of the fourth embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0008] Hereinafter, embodiments will be described with reference to the drawings.

[0009] (First embodiment) 1 is a diagram showing an example of the configuration of a spoofing detection system 1 according to a first embodiment. The spoofing detection system 1 is a system for detecting spoofing of another person by a person passing through a predetermined passage. The predetermined passage is any passage that is subject to spoofing detection, and may be any passage, for example, an entrance / exit to a specific facility, a ticket gate for a transportation facility, etc. The specific facility may be, for example, a company, a medical facility, etc.

[0010] The spoofing detection system 1 illustrated here includes an imaging device 11, an information processing device 12, and an alarm device 13.

[0011] The imaging device 11 is a device for acquiring an image (including a moving image) of the passageway, and may be, for example, a digital camera configured using an imaging element such as a CCD or a CMOS.

[0012] The information processing device 12 is a device that executes a process for detecting impersonation of a person who has entered a passageway based on an image captured by the imaging device 11. The information processing device 12 illustrated here has a configuration similar to that of a general-purpose computer, and includes a calculation device 21, a storage device 22, a communication I / F (Interface) 23, and a user I / F 24. The calculation device 21 is a device that executes predetermined calculation processes and control processes according to a predetermined program, and may be configured using, for example, a CPU (Central Processing Unit), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), etc. The storage device 22 is a device that stores programs and various data required to realize various functions in the impersonation detection system 1, and may be configured using an appropriate non-volatile memory or volatile memory. The communication I / F 23 is a device that establishes communication with external devices such as the imaging device 11 and the alarm device 13 via an appropriate communication network. The user I / F 24 is a device that enables transmission and reception of information between the information processing device 12 and a user (such as an administrator of the spoofing detection system 1), and may be, for example, a display, a keyboard, a pointing device, a touch panel mechanism, a speaker, a microphone, etc.

[0013] The alarm device 13 is a device that notifies the occurrence of impersonation based on the result of processing by the information processing device 12. The specific embodiment of the alarm device 13 should not be particularly limited, but it may be configured, for example, by using a speaker that outputs an alarm sound, a light-emitting device that emits light for notification, a display that displays an image for notification, or the like.

[0014] It should be noted that the above configuration is merely an example, and the configuration of the spoofing detection system 1 is not limited to the above.

[0015] 2 to 4 show an example of a situation in which the spoofing detection system 1 of the present embodiment is applied. Fig. 2 is a side view showing an example of a situation in which the spoofing detection system 1 of the first embodiment is applied. Fig. 3 is a front view showing an example of a situation in which the spoofing detection system 1 of the first embodiment is applied. Fig. 4 is a top view showing an example of a situation in which the spoofing detection system 1 of the first embodiment is applied.

[0016] 2 to 4, passage 35, which is the target of detection by spoofing detection system 1 of this embodiment, is a space formed between two rectangular parallelepiped housings 31A and 31B that are installed parallel to each other. 2 to 4 show an example of a situation in which a person P is about to pass through passage 35 in a traveling direction X while holding up an imitation 40 such as a photograph of another person's face in front of his or her face F.

[0017] The imaging device 11 of this embodiment is installed so as to be able to image the person P who has entered the passage 35 from diagonally forward. In other words, as shown in Fig. 4, the imaging device 11 is installed so that at least the upper body of the person P who has entered the passage 35 is included in its imaging area A and so that its imaging direction Y intersects with the traveling direction X of the passage 35. The imaging device 11 exemplified here is installed near the front end on one of the housings 31A, but the installation position of the imaging device 11 is not limited thereto, and may be any position that allows the person P who has entered the passage 35 to be imaged from diagonally forward.

[0018] Fig. 5 is a diagram showing an example of the functional configuration of the spoofing detection system 1 of the first embodiment. The spoofing detection system 1 of the present embodiment includes an acquisition unit 101, a face detection unit 102, a determination unit 103, an output unit 104, a setting unit 105, and an authentication unit 106. These functional units may be configured by cooperation between hardware elements and software elements as exemplified in Fig. 1. Furthermore, at least one of these functional units may be configured by dedicated hardware (circuitry, etc.).

[0019] The acquisition unit 101 acquires a first captured image (a captured image captured by the imaging device 11) obtained by capturing an image of the passage 35 from diagonally ahead. At this time, when a person P is entering the passage 35, the first captured image is an image of the person P captured from diagonally ahead.

[0020] The face detection unit 102 detects the face of a person present in the first captured image acquired by the acquisition unit 101 .

[0021] The determination unit 103 determines that spoofing is occurring when two or more faces are detected in one first captured image and the distance between a first position corresponding to one of the two faces in the first captured image and a second position corresponding to the other of the two faces is smaller than a predetermined threshold. The threshold may be, for example, a value obtained by converting the thickness of a human body (e.g., a head) in the front-rear direction into a distance on the first captured image.

[0022] When the determination unit 103 determines that spoofing is occurring, the output unit 104 outputs an alarm to notify the occurrence of spoofing. The manner of the alarm is not particularly limited, but may be, for example, the output of an alarm sound, the illumination of a warning light, the display of an alarm screen, or the like.

[0023] The setting unit 105 sets a threshold value used in the determination unit 103 to detect spoofing. The specific method of setting the threshold value is not particularly limited, but for example, a correspondence relationship between the actual distance on the passage 35 and the distance on the first captured image may be calculated based on a preparatory image prepared in advance, and the threshold value may be set based on the correspondence relationship. The preparatory image may be, for example, an image of the passage 35 in a state where a member with a lattice pattern is installed in the passage 35. In this case, the correspondence relationship between the actual distance and the distance on the first captured image can be calculated based on the known dimensions of the lattice pattern. Note that the method of setting the threshold value is not limited to the above, and for example, the threshold value may be set so that the user can arbitrarily change the setting depending on the usage situation (for example, the installation position of the imaging device 11, etc.).

[0024] The authentication unit 106 performs an authentication process to determine whether or not a person corresponding to a face detected by the face detection unit 102 has the authority to pass through the passage 35. The specific method of the authentication process should not be particularly limited, and may be realized by appropriately using a publicly known or new technology. If the result of the authentication process indicates that the person does not have the authority to pass through the passage 35, for example, a process such as closing a door (not shown) that opens and closes the passage 35 may be performed. Furthermore, if the determination unit 103 determines that spoofing is occurring, the authentication unit 106 of this embodiment prohibits the passage of the person entering the passage 35, regardless of the result of the authentication process.

[0025] Fig. 6 is a diagram showing an example of a first captured image I1 in the first embodiment. Fig. 6 illustrates the first captured image I1 capturing a state in which a person P who has entered a passage 35 holds up a replica 40 such as a photograph of another person's face in front of his or her face F.

[0026] In this way, when face detection processing is performed on the first captured image I1, two faces, the real face F of the person P and the imitation 40, are detected. Then, if the distance d between the first position α1 corresponding to the real face F and the second position α2 corresponding to the imitation 40 is smaller than a threshold, it is determined that impersonation is occurring. At this time, the first position α1 may be, for example, the center position of a rectangular first face area β1 set to surround the face F on the first captured image I1. The second position α2 may be, for example, the center position of a rectangular second face area β2 set to surround the imitation 40 on the first captured image I1. The distance d may be calculated from the coordinates of the first position α1 and the coordinates of the second position α2.

[0027] 7 is a diagram showing an example of a method for calculating the coordinates of the first position α1 and the second position α2 in the first embodiment. In FIG. 7, it is shown that the upper left coordinates (coordinates of the upper left point on the first captured image I1) of the first face area β1 are (X1, Y1), and the lower right coordinates (coordinates of the lower right point on the first captured image I1) of the first face area β1 are (X2, Y2). It is also shown that the upper left coordinates of the second face area β2 are (X3, Y3), and the lower right coordinates of the second face area β2 are (X4, Y4). In this case, the coordinates (center coordinates) of the first position α1 corresponding to the face F are [{(X1+X2) / 2}, {(Y1+Y2) / 2)}], and the coordinates of the second position α2 corresponding to the pseudo object 40 are [{(X3+X4) / 2}, {(Y3+Y4) / 2)}].

[0028] An example of a method for setting the threshold value will be described below. As described above, when setting the threshold value, for example, a correspondence relationship between an actual distance on the passage 35 and a distance in the first captured image is calculated based on a preparatory image prepared in advance, and the threshold value can be set based on the correspondence relationship. The calculation of the correspondence relationship is performed using, for example, internal parameters of the imaging device 11.

[0029] Fig. 8 is a diagram showing an example of a situation when a preparatory image used when setting a threshold value in the first embodiment is acquired. As shown in Fig. 8, a checkerboard 41 with a checkered pattern is placed on a passage 35, and the passage 35 on which the checkerboard 41 is installed is imaged by the imaging device 11, whereby the preparatory image can be acquired. At this time, it is assumed that the dimensions of the lattices of the checkerboard 41 used (such as the length of each side of the lattice) are known.

[0030] 9 is a flowchart showing an example of a process for setting a threshold value in the first embodiment. First, a checkerboard 41 is placed on the passage 35 (S101), and a preparatory image of the passage 35 is acquired by the imaging device 11 (S102). After that, the setting unit 105 calculates a correspondence relationship between an actual distance on the passage 35 and a distance on the first captured image by a process such as calibration performed using the preparatory image (S103). Then, the setting unit 105 sets a threshold value based on the correspondence relationship so that the threshold value corresponds to a thickness of a human body in the front-rear direction (S104).

[0031] The method of setting the threshold is not limited to the above. For example, a method may be used in which multiple thresholds are prepared, experiments are performed, and the threshold with the highest spoofing detection rate is selected. Alternatively, a method may be used in which the threshold with the lowest false positive rate is selected, or an optimal threshold may be selected using other appropriate evaluation indices.

[0032] 10 is a flowchart showing an example of processing in the spoofing detection system 1 of the first embodiment. When the acquisition unit 101 acquires the first captured image I1 (S201), the face detection unit 102 executes face detection processing on the first captured image I1 (S202).

[0033] If the result of the face detection process is that no face is present in the first captured image I1 (S203: No), this routine ends. If a face is present in the first captured image I1 (S203: Yes), the determination unit 103 determines whether or not two or more faces are present in the first captured image I1 (S204). If two or more faces are not present in the first captured image I1 (S204: No), that is, if only one face is present in the first captured image I1, the authentication unit 106 executes authentication processing based on the face detected in the first captured image I1 (S205).

[0034] If two or more faces are present in the first captured image I1 (S204: Yes), the determination unit 103 determines whether or not the distance d between the first position α1 corresponding to one of the two faces and the second position α2 corresponding to the other of the two faces is smaller than a threshold (S206). If the distance d is not smaller than the threshold (S206: No), the authentication unit 106 executes authentication processing for each of the two faces (S205).

[0035] If the distance d is smaller than the threshold value (S206: Yes), the determination unit 103 determines that impersonation has occurred (S207), the output unit 104 outputs an alarm, and the authentication unit 106 prohibits the passage of person P entering the passage 35 (S208).

[0036] As described above, according to this embodiment, when two or more faces are detected in the first captured image I1 captured diagonally from the front of the passage 35, it is determined that spoofing is occurring if the distance d between the first position α1 corresponding to one of the two faces in the first captured image I1 and the second position α2 corresponding to the other of the two faces is smaller than a threshold value. This makes it possible to detect spoofing based on an image captured by one imaging device 11, and makes it possible to detect spoofing with high accuracy without increasing costs or complicating the system.

[0037] Other embodiments will be described below with reference to the drawings, but descriptions of parts that are the same as or similar to the first embodiment will be omitted as appropriate.

[0038] Second embodiment The spoofing detection system of the second embodiment differs from the first embodiment in that the authentication process is performed using a second captured image captured from a different direction than the first captured image I1.

[0039] 11 is a diagram showing an example of a functional configuration of the spoofing detection system 2 of the second embodiment. An acquisition unit 201 in the spoofing detection system 2 of this embodiment acquires a first captured image and a second captured image. The first captured image is an image of the person P who has entered the passage 35 captured from an obliquely forward direction, similar to the first captured image I1 in the first embodiment. The second captured image is an image of the person P who has entered the passage 35 captured from a direction different from the first captured image (e.g., directly in front, etc.).

[0040] Furthermore, the face detection unit 202 in the spoofing detection system 2 of this embodiment detects not only faces present in the first captured image but also faces present in the second captured image. Furthermore, when the determination unit 103 determines that spoofing is not occurring, the authentication unit 203 in the spoofing detection system 2 of this embodiment performs authentication processing based on the detection result of the face detection unit 202 performed on the second captured image.

[0041] Fig. 12 is a top view showing an example of a situation in which the spoofing detection system 2 of the second embodiment is applied. As shown in Fig. 12, in this embodiment, an imaging device 51 capable of acquiring a second captured image in which a person P who has entered the passage 35 is captured from directly in front is installed. The imaging device 51 is installed so that its imaging direction Z is parallel or approximately parallel to the traveling direction X of the passage 35. Note that the installation position of the imaging device 51 for acquiring the second captured image is not limited thereto, and it may be any position in which the face F of the person P can be captured from a more frontal side than the imaging device 11 for acquiring the first captured image.

[0042] 13 is a flowchart showing an example of processing in the spoofing detection system 2 of the second embodiment. When the acquisition unit 201 acquires a first captured image I1 (a captured image acquired by the imaging device 11) (S301), the face detection unit 202 executes face detection processing on the first captured image I1 (S302). If the result of the face detection processing shows that a face does not exist in the first captured image I1 (S303: No), this routine ends. If a face exists in the first captured image I1 (S303: Yes), the determination unit 103 determines whether or not two or more faces exist in the first captured image I1 (S304).

[0043] If two or more faces are not present in the first captured image I1 (S304: No), that is, if only one face is present in the first captured image I1, the acquisition unit 201 acquires a second captured image (an image acquired by the imaging device 51) (S306), the face detection unit 202 performs face detection processing on the second captured image (S307), and the authentication unit 203 performs authentication processing based on the face detected in the second captured image (S308).

[0044] If two or more faces are present in the first captured image I1 (S304: Yes), the determination unit 103 determines whether or not the distance d between the first position α1 corresponding to one of the two faces and the second position α2 corresponding to the other of the two faces is smaller than a threshold (S305). If the distance d is not smaller than the threshold (S305: No), the process proceeds to step S306 and subsequent steps.

[0045] If the distance d is smaller than the threshold value (S305: Yes), the determination unit 103 determines that impersonation has occurred (S309), the output unit 104 outputs an alarm, and the authentication unit 203 prohibits the passage of person P entering the passage 35 (S310).

[0046] As described above, according to this embodiment, spoofing detection is performed based on the first captured image in which the person P is captured from an oblique front, and authentication processing is performed based on the second captured image in which the person P is captured from directly in front or substantially directly in front. This makes it possible to perform both spoofing detection and authentication processing with high accuracy.

[0047] Third embodiment The spoofing detection system of the third embodiment differs from the first embodiment in that processing for detecting spoofing is executed only when the entry of a person P into the passage 35 is detected.

[0048] 14 is a diagram showing an example of a functional configuration of the spoofing detection system 3 of the third embodiment. The spoofing detection system 3 of this embodiment includes an intrusion detection unit 301 in addition to the functional units (the acquisition unit 101, the face detection unit 102, the determination unit 103, the output unit 104, and the authentication unit 106) included in the spoofing detection system 1 of the first embodiment.

[0049] The entry detection unit 301 detects that a person P has entered the passage 35. The specific configuration of the entry detection unit 301 is not particularly limited, and may be configured using an appropriate human sensor, for example, an infrared sensor or the like installed in the housings 31A and 31B. Also, the entry of the person P into the passage 35 may be detected by image analysis of an appropriate captured image (for example, the first captured image I1).

[0050] 15 is a flowchart showing an example of processing in the spoofing detection system 3 of the third embodiment. After the acquisition unit 101 acquires the first captured image I1 (S401), the entry detection unit 301 determines whether or not the person P has entered the passage 35 (S402). If the entry of the person P is not detected (S402: No), this routine ends. Note that step S402 may be executed before step S401. In this case, the first captured image I1 is acquired when the entry of the person P is detected.

[0051] If the intrusion of a person P is detected (S402: Yes), the face detection unit 102 executes face detection processing on the first captured image I1 (S403). If the result of the face detection processing is that no face is present in the first captured image I1 (S404: No), this routine ends. If a face is present in the first captured image I1 (S404: Yes), the determination unit 103 determines whether or not two or more faces are present in the first captured image I1 (S405). If two or more faces are not present in the first captured image I1 (S405: No), that is, if only one face is present in the first captured image I1, the authentication unit 106 executes authentication processing based on the face detected in the first captured image I1 (S406).

[0052] If two or more faces are present in the first captured image I1 (S405: Yes), the determination unit 103 determines whether the distance d between the first position α1 corresponding to one of the two faces and the second position α2 corresponding to the other of the two faces is smaller than a threshold (S407). If the distance d is not smaller than the threshold (S408: No), the authentication unit 106 executes authentication processing for each of the two faces (S406).

[0053] If the distance d is smaller than the threshold value (S407: Yes), the determination unit 103 determines that impersonation has occurred (S408), the output unit 104 outputs an alarm, and the authentication unit 106 prohibits the person P entering the passage 35 from passing through (S409).

[0054] As described above, according to this embodiment, the face detection process, the impersonation detection process, and the authentication process are performed only when the person P enters the passage 35, so that the processing load on the information processing device 12 can be reduced.

[0055] (Fourth embodiment) The impersonation detection system of the fourth embodiment differs from the first embodiment in that after person P enters the passage 35, the position and height of person P are calculated based on the results of face detection processing on the first captured image I1, and the threshold value used for detecting impersonation is corrected based on the calculated position and height.

[0056] 16 is a diagram showing an example of a functional configuration of the spoofing detection system 4 of the fourth embodiment. The spoofing detection system 3 of this embodiment includes an intrusion detection unit 301 and a correction unit 302 in addition to the functional units (acquisition unit 101, face detection unit 102, determination unit 103, output unit 104, and authentication unit 106) included in the spoofing detection system 1 of the first embodiment.

[0057] The entry detection unit 301 detects that a person P has entered the passage 35, similarly to the third embodiment.

[0058] After the entry of the person P into the passage 35 is detected by the entry detection unit 301, the correction unit 302 calculates the standing position and height of the person P based on the result of face detection processing on the first captured image by the face detection unit 102. The height may be, for example, the position of the center point of the face of the person P or the height. The standing position may be, for example, the position of a point used in calculating the height (for example, the center point of the face, etc.) projected onto the floor. Then, the correction unit 302 corrects the threshold value used for detecting impersonation in the determination unit 103 based on the calculated standing position and height. The correspondence relationship between the actual distance on the passage 35 and the distance on the first captured image changes depending on the standing position and height of the person P, but the correction unit 302 corrects the threshold value so as to reduce the influence of such changes in the standing position and height of the person P.

[0059] 17 is a flowchart showing an example of processing in the spoofing detection system 4 of the fourth embodiment. After the acquisition unit 101 acquires the first captured image I1 (S501), the entry detection unit 301 determines whether or not the person P has entered the passage 35 (S502). If the entry of the person P is not detected (S502: No), this routine ends. Note that step S502 may be executed before step S501. In this case, the first captured image I1 is acquired when the entry of the person P is detected.

[0060] If the entry of the person P is detected (S502: Yes), the face detection unit 102 executes face detection processing on the first captured image I1 (S503). After that, the correction unit 302 calculates the standing position and height of the person P who entered the passage 35 based on the result of the face detection processing (S504), and corrects the threshold value used for detecting impersonation based on the calculated standing position and height (S505).

[0061] After that, the determination unit 103 determines whether or not two or more faces are present in the first captured image I1 (S506). If two or more faces are not present in the first captured image I1 (S506: No), that is, if only one face is present in the first captured image I1, the authentication unit 106 executes authentication processing based on the face detected in the first captured image I1 (S507).

[0062] If two or more faces are present in the first captured image I1 (S506: Yes), the determination unit 103 determines whether the distance d between the first position α1 corresponding to one of the two faces and the second position α2 corresponding to the other of the two faces is smaller than the corrected threshold (S508). If the distance d is not smaller than the threshold (S508: No), the authentication unit 106 executes authentication processing for each of the two faces (S507).

[0063] If the distance d is smaller than the threshold value (S508: Yes), the determination unit 103 determines that impersonation has occurred (S509), the output unit 104 outputs an alarm, and the authentication unit 106 prohibits the passage of person P entering the passage 35 (S510).

[0064] As described above, according to this embodiment, the threshold value can be optimized according to the position, physique, etc. of the person P who has entered the passage 35, and the accuracy of detecting impersonation can be improved.

[0065] A program for causing a computer (such as the information processing device 12) to execute processes for realizing the functions of the spoofing detection systems 1 to 3 as described above is provided by being recorded in a computer-readable recording medium such as a CD-ROM, a flexible disk (FD), a CD-R, or a digital versatile disk (DVD) in an installable or executable format file. The program may be provided by being stored on a computer connected to a network such as the Internet and downloaded via the network. The program may be provided or distributed via a network such as the Internet. The program may be provided by being pre-installed in a ROM or the like.

[0066] Although the embodiments of the present invention have been described above, the embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their modifications are included in the scope and spirit of the invention, and are included in the scope of the invention and its equivalents described in the claims. [Explanation of symbols]

[0067] 1 to 4: spoofing detection system, 11, 51: imaging device, 31A, 31B: housing, 35: passageway, 40: imitation object, 41: checkerboard, 101, 201: acquisition unit, 102, 202: face detection unit, 103: determination unit, 104: output unit, 105: setting unit, 106, 203: authentication unit, 301: entry detection unit, 302: correction unit, d: distance, F: face, I1: first captured image, P: person, α1: first position, α2: second position, β1: first face area, β2: second face area

Claims

1. An impersonation detection system for detecting an impersonation of another person by a person passing through a predetermined passage, comprising: an acquisition unit that acquires a first captured image obtained by capturing an image of a person who has entered the passage from an oblique front; a face detection unit that detects a face present in the first captured image; a determination unit that determines that the spoofing is occurring when two or more faces are detected in one of the first captured images and a distance between a first position corresponding to one of the two faces in the first captured image and a second position corresponding to the other of the two faces is smaller than a predetermined threshold; A spoofing detection system comprising:

2. The threshold value is a value obtained by converting a thickness of a human body in a front-back direction into a distance on the first captured image. The spoofing detection system of claim 1 .

3. The threshold value is a value obtained by converting the thickness of the head in the front-rear direction into a distance on the first captured image. The spoofing detection system of claim 1 .

4. a setting unit that calculates a correspondence relationship between an actual distance on the passage and a distance on the first captured image based on a preparatory image obtained by capturing an image of the passage in a state in which a member having a lattice pattern is placed on the passage, and sets the threshold value based on the correspondence relationship; The spoofing detection system of claim 1 further comprising:

5. an authentication unit that performs an authentication process to determine whether or not a person corresponding to a face detected by the face detection unit has authority to pass through the passage, and when the determination unit determines that impersonation is occurring, prohibits the person from passing through the passage; The spoofing detection system of claim 1 further comprising:

6. the acquisition unit acquires a second captured image obtained by capturing an image of the person who has entered the passage from a direction different from an imaging direction of the first captured image, The face detection unit detects a face present in the second captured image, When the determination unit determines that the spoofing is not occurring, the authentication unit performs the authentication process based on a detection result of the face detection unit performed on the second captured image. The spoofing detection system of claim 5.

7. an entrance detection unit that detects when a person enters the passage; a correction unit that, after the entry of a person into the passage is detected, calculates a standing position and a height of the person who has entered the passage based on a detection result of the face detection unit, and corrects the threshold value based on the standing position and the height; The spoofing detection system of claim 1 further comprising:

8. the first position is a center position of a rectangular area set to surround the one face in the first captured image, the second position is a center position of a rectangular area set to surround the other face in the first captured image; The spoofing detection system of claim 1 .

9. 1. A method for detecting impersonation of another person by a person passing through a predetermined passage, comprising the steps of: acquiring a first captured image of the person entering the passage from an obliquely forward position; detecting a face present in the first captured image; a step of determining that the spoofing is occurring when two or more faces are detected in one of the first captured images and a distance between a first position corresponding to one of the two faces in the first captured image and a second position corresponding to the other of the two faces is smaller than a predetermined threshold; A spoofing detection method comprising:

10. A computer that executes a process for detecting impersonation of another person by a person passing through a predetermined passageway, A process of acquiring a first captured image of a person who has entered the passage from an obliquely forward direction; A process of detecting a face present in the first captured image; a process of determining that the spoofing is occurring when two or more faces are detected in one of the first captured images, and a distance between a first position corresponding to one of the two faces in the first captured image and a second position corresponding to the other of the two faces is smaller than a predetermined threshold; A program that executes the following.

Images