Access control system, access control method, and program

Abstract

To reduce the possibility that a person who first entered a passageway after authentication may pass through the passageway when a person who was authenticated before entering the passageway and the person who first entered the passageway after authentication are different individuals, at a ticket gate, etc. at train stations.SOLUTION: In a passage control system according to the embodiment, a control device performs the processing of: inputting authentication information for a first person before entering a passageway; verifying whether the first person is an authorized passerby based on the authentication information; detecting the first person's face based on a captured image when the first person is verified as the authorized passerby; tracking the first person's face based on multiple captured images in sequence; setting an image region based on a detected position of a second person who first entered the passageway after the first person was verified as the authorized user; determining whether the face of the tracked first person is within the image region in the captured image when the image region is set; and performing control to close an opening / closing door when determined to be not within the image region.SELECTED DRAWING: Figure 7

Description

[Technical Field]

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

[0002] Conventionally, passage control has been performed at ticket gates at railway stations, airports, etc., and at entrances and exits of important areas, etc. Below, we will take the case of a ticket gate at a railway station as an example. A ticket gate at a railway station includes, for example, a housing that forms a passageway for passengers to pass through, a card reader provided at the entrance of the housing, and an opening / closing door provided at the exit of the housing.

[0003] Then, for example, the user holds an IC (Integrated Circuit) card (such as PASMO (registered trademark)) over a card reader at the entrance of the ticket gate, and if the authentication result is successful, the door at the exit opens and the user can pass through the passageway. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2022-164659 [Patent Document 2] Japanese Patent Application Laid-Open No. 2015-114917 [Patent Document 3] Japanese Patent Publication No. 2022-138548 [Patent Document 4] Japanese Patent Application Laid-Open No. 2004-265232 Summary of the Invention [Problem to be solved by the invention]

[0005] However, in the case of the above-mentioned conventional technology, if the authentication result of the user's IC card is good, another person can cut in line and go through the passage before the user. However, the conventional technology does not have a sufficient means to prevent this.

[0006] Therefore, an object of an embodiment of the present invention is to provide a passage control system, a passage control method, and a program that can reduce the possibility that a person who is authenticated before entering an aisle and a person who first enters the aisle after authentication will pass through the aisle when the latter person is different from the latter person during passage control. [Means for solving the problem]

[0007] The passage control system of the embodiment is a passage control system including a housing forming a passage for people to pass through, an opening / closing door provided at an exit portion of the housing, a photographing unit that photographs the passage and an area in front of the passage and outputs the photographed image, a detection unit that detects the position of a person who has entered the passage, and a control device, wherein the control device includes an authentication information input unit that inputs authentication information of a first person before entering the passage, a verification unit that verifies whether the first person is an authorized passerby based on the authentication information, and a control device that verifies whether the first person is an authorized passerby based on the photographed image when the verification unit has verified that the first person is an authorized passerby. The system includes a detection unit that detects a person's face, a tracking unit that tracks the face of the first person based on the plurality of captured images in chronological order, an image area setting unit that sets an image area according to the detection position of a second person who enters the passageway for the first time when the second person is detected by the detection unit after the first person is matched by the matching unit as a legitimate passerby, a determination unit that determines whether the face of the first person being tracked by the tracking unit is within the image area in the captured images when the image area is set by the image area setting unit, and a control unit that controls the opening and closing of the door when the determination unit determines that the face is not within the image area. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a diagram illustrating a typical configuration of a traffic control system according to an embodiment at a site. [Figure 2] FIG. 2 is a block diagram showing the functional configuration of the traffic control system according to the embodiment. [Figure 3] FIG. 3 is a diagram showing an example of a face detection result in the embodiment. [Figure 4] FIG. 4 is a diagram showing an example of a face tracking result in the embodiment. [Figure 5] FIG. 5 is a diagram illustrating an example of image region information according to the embodiment. [Figure 6] FIG. 6 is a diagram showing an example of setting an image area in the embodiment. [Figure 7] FIG. 7 is a flowchart showing the processing performed by the traffic control system according to the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, embodiments of a traffic control system, a traffic control method, and a program according to the present invention will be described with reference to the drawings. In this embodiment, a ticket gate at a railway station will be used as an example. In the following, an image captured by the image capture unit 3 (FIGS. 1 and 2) may be referred to as an "image" or an "input image."

[0010] Fig. 1 is a diagram showing a schematic configuration of an on-site passage control system S according to an embodiment. Fig. 1(a) is a diagram showing a side view of a ticket gate housing 1, and Fig. 1(b) is a diagram showing a top view of the ticket gate housing 1. The passage control system S includes the housing 1, an opening / closing door 2, a photographing unit 3, a card reader 4, a detection unit 5, a control device 6, etc.

[0011] The pair of housings 1 are members that form a passageway for a person M to pass through. The pair of opening / closing doors 2 are provided at the exit portion of the housing 1, and are opened by an open signal from the control device 6, and are closed by a close signal from the control device 6.

[0012] The photographing unit 3 photographs the passageway and the area in front of the passageway, and transmits the photographed images to the control device 6. The photographing unit 3 is, for example, an ITV (Industrial Television) camera (monitoring camera). The ITV camera digitizes optical information obtained through a lens using an A / D (Analog to Digital) converter and outputs image data. Furthermore, the photographing unit 3 is not limited to an ITV camera, and may be another camera such as a digital still camera.

[0013] The card reader 4 is provided at the entrance of the housing 1, and reads authentication information (user information) from an IC card 9 (for example, PASMO (registered trademark) or the like) brought close by the person M, and transmits the information to the control device 6.

[0014] The detection unit 5 detects the position of a person who has entered the passage (for example, a position around position B in FIG. 1(b)). For example, the detection unit 5 detects whether a person is present or not at each specific position in the passage. The detection unit 5 is, for example, a human detection sensor (motion sensor). The detection unit 5 is installed, for example, at the entrance or middle of the passage so that it is easy to detect that a person has entered the passage. Furthermore, the passage may be divided into several areas, and multiple detection units 5 may be installed so as to detect the presence or absence of a person in each area. In other words, although the detection unit 5 is depicted as being one in FIGS. 1 and 2, the number of detection units 5 may be two or more.

[0015] In this embodiment, the "first person" refers to the person who operates the IC card 9 in the card reader 4. Furthermore, the "second person" refers to the person (who may be the first person, or another person who cuts in line) who is first detected in the passage by the detection unit 5 after the first person is verified as a legitimate passerby by the verification unit 653 (described later).

[0016] Next, the control device 6 and the like will be described with reference to Fig. 2. Fig. 2 is a block diagram showing the functional configuration of the traffic control system S of the embodiment. The control device 6 includes a storage unit 64 and a processing unit 65.

[0017] The storage unit 64 is configured by a RAM (Random Access Memory), a ROM (Read Only Memory), an SSD (Solid State Drive), etc., and stores various information (programs, setting data, calculation results, etc.).

[0018] The processing unit 65 is configured by, for example, a CPU (Central Processing Unit), and by the CPU executing a program stored in the storage unit 64, the processing unit 65 realizes, as functional components, an authentication information input unit 61, an image input unit 62, a detection information input unit 63, a detection unit 651, a tracking unit 652, a matching unit 653, an image area setting unit 654, a determination unit 655, and a control unit 656. Note that some or all of the functional components may be realized by hardware such as an FPGA (Field Programmable Gate Array) or an ASIC (Application Specific Integrated Circuit).

[0019] The authentication information input unit 61 inputs authentication information of a first person before entering the passage. For example, the authentication information input unit 61 inputs authentication information from the card reader 4. Note that the authentication information input unit 61 only needs to be able to input authentication information of a person before entering the passage, and the information to be input is not limited to authentication information from the card reader 4, but may also be authentication information from a magnetic ticket, a barcode, a two-dimensional code, a smartphone, the photographing unit 3, or the like. In this case, the authentication information input unit 61 may input authentication information using radio wave communication such as Bluetooth (registered trademark). The authentication information may also be, for example, information regarding a passage permit, ID (identifier) ​​information of an authorized passerby, a facial image of an authorized passerby, or the like.

[0020] The image input unit 62 inputs the captured image from the image capturing unit 3 .

[0021] The detection information input unit 63 receives a position detection signal of a person who has entered the passage from the detection unit 5 and outputs it to the image area setting unit 654.

[0022] The detection unit 651 performs various detections. For example, the detection unit 651 detects the face of a first person based on the captured image when the matching unit 653 matches the first person as a legitimate pedestrian. For example, the detection unit 651 performs face detection processing to detect the face of a person from multiple captured images input in a time series. The face detection processing can be realized, for example, by creating an average image of rectangular areas of many faces as a face template, scanning the face template while translating, rotating, and scaling it on the image acquired by the image input unit 62, and determining the position where the difference in brightness value with the face template is smallest as the face detection area. Furthermore, the face detection area can be expressed, for example, in the form of coordinates on the image of the upper left and lower right of a rectangle surrounding the face.

[0023] 3 is a diagram showing an example of a face detection result in this embodiment. In this example, in addition to the coordinates of the face rectangle, a face rectangle ID is also included to distinguish between multiple faces in the image. Returning to FIG. 2, the detection unit 651 outputs the face detection result to the tracking unit 652.

[0024] The tracking unit 652 tracks the face of the first person based on multiple captured images in chronological order. For example, the tracking unit 652 performs a tracking process to associate the face of the same person between frames with the face detected by the detection unit 651 from the input image of each frame. This tracking process can be realized, for example, by using a method of associating the position of a detected face in the next frame using optical flow, or the method described in Japanese Patent Application Laid-Open No. 2011-170711.

[0025] 4 is a diagram showing an example of a face tracking result in an embodiment. In this example, the face tracking result includes the frame number of a frame in which a face is present, the coordinates of a face rectangle in the image of that frame, and the ID of that face rectangle. The same ID is assigned to faces that are considered to be the same person in previous and subsequent frames.

[0026] Returning to Fig. 2, the matching unit 653 checks whether the first person is an authorized passerby based on the authentication information received from the authentication information input unit 61. For example, the matching unit 653 performs processing to determine the facial similarity between the face obtained by each face tracking and the face in the registered image associated with the personal ID input by the authentication information input unit 61. The processing to determine the facial similarity can be realized, for example, as in the method disclosed in Japanese Patent Application Laid-Open No. 2007-4767, by using a model to create an image in which the orientation and state of the face are intentionally changed for a single piece of person image information, and then obtaining facial feature information and performing matching.

[0027] Alternatively, a method described in a literature (Kazuhiro Fukui, Osamu Yamaguchi, Kenichi Maeda, "Facial Recognition System Using Moving Images," Institute of Electronics, Information and Communication Engineers Research Report PRMU, Vol. 97, No. 113, pp. 17-24 (1997)) may be used, which performs calculations using multiple frames of the same person in a moving image. By using feature information such as a subspace output by such a method as feature information for an individual face detected in an input image, it becomes possible to realize processing for determining facial similarity. In this case, the similarity index may be, for example, the similarity between subspaces managed as facial feature information.

[0028] In addition, similarity indices other than similarity, such as distance in a feature space or Mahalanobis distance, can also be used. When distance is used, the larger the value, the lower the similarity. Therefore, when comparing with a threshold, the smaller the value is, the higher the similarity to the registered person. When the obtained similarity is equal to or greater than a predetermined threshold, the detected face and the person being compared are determined to be the same person, and when it is less than the threshold, they are determined to be different people. In addition, the registered image associated with the personal ID can be obtained, for example, from those stored in a database of registered images using the personal ID as a key. In addition, a facial image of an authorized pedestrian input via the authentication information input unit 61 can also be used.

[0029] Furthermore, as a process of the matching unit 653, for example, face matching may be performed against the entire database of registered images, and a determination may be made as to whether or not the person is a legitimate passerby based on whether or not the personal ID corresponding to the registered image that exceeds the threshold matches the personal ID inputted by the authentication information input unit 61.

[0030] The image area setting unit 654 sets an image area according to the detection position of a second person who entered the passageway for the first time when the second person was detected by the detection unit 5 after the matching unit 653 matched the first person to be a regular passerby. For example, the image area setting unit 654 sets an image area where the face of a person at the position detected in the passageway by the detection unit 5 is present, and outputs information about the image area to the determination unit 655.

[0031] For example, in the case where there are multiple detection units 5 and the image area is rectangular, Fig. 5 is a diagram showing an example of image area information in the embodiment. The sensor number is identification information for each detection unit 5. Fig. 6 is a diagram showing an example of setting the image area R in the embodiment. In other words, if a person is present at the position detected by a certain detection unit 5, the image area R is the range in the captured image where the person's face may be present.

[0032] 2, when the image area is set by the image area setting unit 654, the determination unit 655 determines whether or not the face of the first person being tracked by the tracking unit 652 is within the image area R (FIG. 6) in the captured image. For example, the determination unit 655 inputs the face tracking result of the authorized pedestrian matched by the matching unit 653 with the set image area, and determines whether or not the tracked face is within the image area.

[0033] Specifically, for example, if the set image area is rectangular, with the coordinates of the upper left corner being (X1, Y1) and the coordinates of the lower right corner being (X2, Y2), and the coordinates of the upper left corner of the rectangle of the tracked face immediately after detection information is input by the detection unit 5 are (x1, y1) and (x2, y2), then the tracked face is determined to be within the image area if the following four inequalities (1) to (4) hold, and is determined to be not within the area otherwise. Note that, as shown in Fig. 6, the x-axis has a direction in which values ​​increase to the right, and the y-axis has a direction in which values ​​increase downward.

[0034] X1 ≦ x1 (1) Y1 ≦ y1 (2) X2 ≧ x2 (3) Y2 ≧ y2 (4)

[0035] The control unit 656 executes various controls. The control unit 656 controls the opening and closing of the opening and closing door 2. For example, when the determination unit 655 determines that the opening and closing door 2 is not possible, the control unit 656 controls the opening and closing door 2 to close.

[0036] 7 is a flowchart showing the processing by the traffic control system S of the embodiment. First, in step S101, the image input unit 62 inputs a captured image from the image capturing unit 3. Next, in step S102, the detection unit 651 performs face detection processing using the captured image.

[0037] Next, in step S103, the detection unit 651 determines whether or not a face has been detected in the captured image, and if Yes, the process proceeds to step S104, and if No, the process returns to step S101. In step S104, the tracking unit 652 performs face tracking processing.

[0038] Next, in step S105, the authentication information input unit 61 determines whether or not authentication information has been input from the card reader 4, and if Yes, the process proceeds to step S111, and if No, the process proceeds to step S106.

[0039] In step S106, when authentication information is received from the card reader 4 operated on the IC card 9, the authentication information input unit 61 inputs the authentication information.

[0040] Next, in step S107, the authentication information input unit 61 determines whether or not authentication information has been input from the card reader 4, and if Yes, the process proceeds to step S108, and if No, the process returns to step S101. In step S108, the matching unit 653 performs face matching processing.

[0041] Next, in step S109, the matching unit 653 determines whether the first person is a legitimate passerby based on the authentication information and the result of the face matching process, and if Yes, proceeds to step S110, and if No, proceeds to step S116.

[0042] In step S110, the control unit 656 performs processing to open the door 2 (that is, sends an open signal to the door 2). This opens the door 2. Thereafter, the process returns to step S101, and step S105 becomes Yes, so the process proceeds to step S111.

[0043] In step S111, the detection information input unit 63 performs an input process of the detection information from the detection unit 5. Next, in step S112, the detection information input unit 63 determines whether or not detection information has been input from the detection unit 5. If Yes, the process proceeds to step S113, and if No, the process returns to step S101.

[0044] In step S113, the image area setting unit 654 sets an image area R (FIG. 6) in the captured image according to the detected position of the person in the detection information.

[0045] Next, in step S114, the determination unit 655 determines whether or not the first person is a legitimate pedestrian, that is, whether or not the face of the first person being tracked by the tracking unit 652 is in the image area R (FIG. 6) in the captured image.

[0046] Next, in step S115, the determination unit 655 determines whether or not the person is a regular passerby, and if Yes, the process ends, and if No, the process proceeds to step S116.

[0047] In step S116, the control unit 656 performs a process to close the door 2 (that is, sends a close signal to the door 2). As a result, the door 2 closes.

[0048] In this way, according to the passage control system S of this embodiment, in passage control using a ticket gate, it is possible to determine with high accuracy whether the person who was authenticated before entering the passage is different from the person who first enters the passage after authentication, and if these people are different, the opening and closing door 2 is closed, thereby preventing the latter person from passing through the passage.

[0049] Furthermore, authentication before entering the passage can be performed using authentication information read from the IC card 9 by the card reader 4. This allows the present invention to be easily applied to ticket gates at railway stations and the like that are normally operated in this way.

[0050] Here, a comparison with conventional technologies will be explained. For example, a first conventional technology is a method that uses cameras and sensors to detect the entry of unauthorized passersby before authorized passersby who have been authenticated enter an area. However, with this method, a person behind an authorized passerby is determined to be an unauthorized passerby at the time of authentication, but there is a problem in that it cannot detect the intrusion of an unauthorized passerby when the unauthorized passerby approaches the authorized passerby and it is not clear which passerby is in front of them, or when an authorized passerby intentionally gets in front of an unauthorized passerby.

[0051] Furthermore, for example, as a second conventional technique, when performing facial matching of people passing through a passageway, when there are multiple people captured on camera and the subjects for facial matching are narrowed down, for a person who has already entered the passageway and is not a subject for facial matching, information from a human detection sensor is used to identify the position of that person in the passageway, and a face in an image that is estimated from the identified position is excluded from the subjects for matching. However, because a second matching is not performed when entering the passageway, there is a problem that if an authorized passerby is authenticated before entering the passageway, and then an unauthorized passerby enters the passageway before the authorized passerby enters the passageway, the unauthorized passerby can pass through.

[0052] On the other hand, with the traffic control system S of this embodiment, the possibility of such problems occurring is low.

[0053] A program that causes the processing unit 65 (computer) of the control device 6 to execute processing for realizing the functions of the above-described embodiment can be provided by being recorded in an installable or executable file format on a computer-readable recording medium such as a CD (Compact Disc)-ROM, a flexible disk (FD), a CD-R (Recordable), or a DVD (Digital Versatile Disk).The program may also be provided or distributed via a network such as the Internet.

[0054] Although the embodiments of the present invention have been described above, they are presented as examples and are not intended to limit the scope of the invention. The novel embodiments described above can be embodied in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. The embodiments and their modifications are included within the scope and spirit of the invention, and are also included in the inventions and their equivalents as defined in the claims.

[0055] For example, although the above embodiment has been described with reference to a ticket gate at a train station, the present invention is not limited to this example and may also be applied to other situations, such as ticket gates at other locations, such as airports, or entrances and exits to important areas (e.g., important areas in companies, medical facilities, etc.).

[0056] The number of the photographing unit 3 is not limited to one, but may be two or more. The installation position of the photographing unit 3 is not limited to the housing 1, but may be other locations such as the ceiling.

[0057] Furthermore, if the face rectangle of an authorized pedestrian is within the image area, and if a face tracking result different from that of the authorized pedestrian is also input, and the face rectangle immediately after the detection information of the tracking result is input satisfies the four inequalities (1) to (4), the determination may be suspended, and determination processing may be performed on the next image input from the image input unit 62. In this case, for example, if the determination suspension continues up to a predetermined upper limit number of times, it may be determined that the face rectangle of an authorized pedestrian is not present within the image area.

[0058] In addition, a detection unit similar to detection unit 5 may be installed at the exit of the passageway, and when that detection unit detects that a person has passed through the exit, it may be used as a trigger to end a series of processes or start the next series of processes. [Explanation of symbols]

[0059] 1...housing, 2...opening / closing door, 3...photographing unit, 4...card reader, 5...detection unit, 6...control device, 9...IC card, 61...authentication information input unit, 62...image input unit, 63...detection information input unit, 64...storage unit, 65...processing unit, 651...detection unit, 652...tracking unit, 653...matching unit, 654...image area setting unit, 655...determination unit, 656...control unit, M...person, S...traffic control system

Claims

1. a housing that forms a passageway for people to pass through; an opening / closing door provided at an exit portion of the housing; an image capturing unit that captures an image of the passageway and an area in front of the passageway and outputs the captured image; a detection unit that detects the position of a person who has entered the passage; A traffic control system comprising: The control device an authentication information input unit for inputting authentication information of a first person before entering the passage; a verification unit that verifies whether the first person is an authorized passerby based on the authentication information; a detection unit that detects a face of the first person based on the captured image when the first person is verified as an authorized pedestrian by the verification unit; a tracking unit that tracks the face of the first person based on the plurality of captured images in time series; an image area setting unit that sets an image area in accordance with a detected position of a second person when the second person is detected by the detection unit for the first time after the first person is verified as an authorized passerby by the verification unit; a determination unit that determines whether or not a face of the first person being tracked by the tracking unit is in the image area in the captured image when the image area is set by the image area setting unit; A control unit that controls the opening and closing of the opening door when the determination unit determines that the condition is not met.

2. The device further includes a card reader provided at an entrance portion of the housing, The passage control system according to claim 1 , wherein the authentication information input unit inputs the authentication information read from a contactless medium brought close by the first person through the card reader.

3. a housing that forms a passageway for people to pass through; an opening / closing door provided at an exit portion of the housing; an image capturing unit that captures an image of the passageway and an area in front of the passageway and outputs the captured image; a detection unit that detects the position of a person who has entered the passage; A traffic control method in a traffic control system including a control device, The control device an authentication information input step of inputting authentication information of a first person before entering the passage; a verification step of verifying whether the first person is an authorized passerby based on the authentication information; a detection step of detecting a face of the first person based on the captured image when the first person is verified as an authorized passerby in the verification step; a tracking step of tracking a face of the first person based on the plurality of captured images in time series; an image area setting step of setting an image area in accordance with a detected position of a second person who enters the passage for the first time by the detection unit after the first person has been verified as a legitimate passerby in the verification step; a determination step of determining whether or not a face of the first person being tracked by the tracking step is in the image area in the captured image when the image area is set by the image area setting step; a control step of controlling the opening and closing of the door when the determination step determines that the condition is not met.

4. a housing that forms a passageway for people to pass through; an opening / closing door provided at an exit portion of the housing; an image capturing unit that captures an image of the passageway and an area in front of the passageway and outputs the captured image; a detection unit that detects the position of a person who has entered the passage; a computer serving as the control device in a traffic control system including the control device; an authentication information input unit for inputting authentication information of a first person before entering the passage; a verification unit that verifies whether the first person is an authorized passerby based on the authentication information; a detection unit that detects a face of the first person based on the captured image when the first person is verified as an authorized pedestrian by the verification unit; a tracking unit that tracks the face of the first person based on the plurality of captured images in time series; an image area setting unit that sets an image area in accordance with a detected position of a second person when the second person is detected by the detection unit for the first time after the first person is verified as an authorized passerby by the verification unit; a determination unit that determines whether or not a face of the first person being tracked by the tracking unit is in the image area in the captured image when the image area is set by the image area setting unit; A program for causing the control unit to function as a control unit that controls the opening and closing of the door when the judgment unit judges that the door is not open.

Images