JP7075702B2 - Entry / exit authentication system and entry / exit authentication method - Google Patents

Description

The present disclosure relates to an entry / exit authentication system for authenticating the opening / closing of an entrance and an entry / exit authentication method.

In Patent Document 1, whether or not the person located at the entrance has the authority to enter the managed area based on the ID information wirelessly acquired from the terminal possessed by the person located at the entrance of the managed area. An ID authentication unit that determines whether or not an ID is authenticated, and a face information acquisition unit that acquires face information that indicates whether or not a face image of a predetermined size is included in a photographed image taken from a predetermined position at the entrance. And an entrance / exit management device having a door control unit that controls an opening / closing state of an entrance based on the result of ID authentication and face information is disclosed.

Japanese Unexamined Patent Publication No. 2017-8638

In Patent Document 1, the acquisition of ID information of an ID card for entering a managed area is provided on the condition that the size occupied by a human face image with respect to the captured image taken by the camera is equal to or larger than a predetermined value. And ID authentication is performed. However, in the configuration of Patent Document 1, the direction of the human face or the direction of the line of sight is not taken into consideration. For this reason, for example, when a person who does not intend to enter or leave the room passes near the entrance of the managed area, if the size occupied by the face image of the person who passed the photographed image is equal to or larger than a predetermined value, ID authentication is performed. Depending on the result, it was possible to enter the controlled area, and it was possible for unwanted people to enter the room, which could lead to deterioration of security.

The present disclosure is devised in view of the above-mentioned conventional situation, and provides an entry / exit authentication system and an entry / exit authentication method for effectively suppressing entry / exit of a person who does not intend to enter / leave the room and improving security. The purpose is.

The present disclosure is an entry / exit authentication system in which a computer, a face authentication device, and an entrance / exit management device are communicably connected, and the computer is a camera that captures an image of a person located near the entrance of a managed area. The setting screen of the entrance and the face detection area of the person including the captured image is displayed, and the entrance and the face detection area are set by the detection based on the captured image of the camera or the user operation, and the face authentication device is used. The face authentication device detects whether or not the person is facing the entrance based on the setting result of the entrance and the face detection area from the computer and the image captured by the camera, and the detection thereof. The result is sent to the entrance / exit management device, and when the entry / exit management device detects that the person is facing the entrance based on the detection result sent from the face authentication device, the person said. Provides an entry / exit authentication system that controls the open / closed state of the entrance according to a determination result of whether or not a person is permitted to enter or leave the controlled area.

Further, the present disclosure is an entry / exit authentication method using an entry / exit authentication system in which a computer, a face recognition device, and an entry / exit management device are communicably connected, and the computer is near the entrance of a managed area. The setting screen of the entrance and the face detection area of the person including the image captured by the camera that captures the person located in is displayed, and the entrance and the face detection area are detected by detection based on the image captured by the camera or by user operation. Is set and sent to the face recognition device, and the face recognition device faces the entrance by the person based on the setting result of the entrance and the face detection area from the computer and the image captured by the camera. Whether or not it is detected, the detection result is sent to the entrance / exit management device, and the person is facing the entrance of the entrance / exit management device based on the detection result sent from the face recognition device. When it is detected, an entry / exit authentication method for controlling the open / closed state of the entrance is provided according to a determination result of whether or not the person is permitted to enter or leave the controlled area. ..

According to the present disclosure, it is possible to effectively suppress the entry and exit of a person who does not intend to enter and leave the room, and it is possible to improve security.

The figure which shows the inside of the facility which installed the entry / exit management system which concerns on Embodiment 1. Diagram showing a configuration example of an entry / exit management system A diagram showing a hardware configuration example of a face authentication server and an entry / exit management server. The figure which shows the setting screen which sets a door area and a face detection area in the image taken from the side of a door. The figure which shows the setting screen which sets a door area and a face detection area in the image taken from the front direction of a door. Sequence diagram showing the setting operation in the entry / exit management system Sequence diagram showing operation during operation in the entry / exit management system A diagram showing the operating status of multiple people passing in front of the door when the angle of view of the face recognition camera is set to wide. A diagram showing the operating status of multiple people passing in front of the door when the angle of view of the face recognition camera is set to a narrow angle. A diagram showing the operating status of multiple people passing in front of a door when two face recognition cameras are installed. A sequence diagram showing an operation during operation in the entry / exit management system of the second embodiment.

Hereinafter, embodiments in which the entry / exit authentication system and the entry / exit authentication method according to the present disclosure are specifically disclosed will be described in detail with reference to the drawings as appropriate. However, more detailed explanation than necessary may be omitted. For example, detailed explanations of already well-known matters and duplicate explanations for substantially the same configuration may be omitted. This is to avoid unnecessary redundancy of the following description and to facilitate the understanding of those skilled in the art. It should be noted that the accompanying drawings and the following description are provided for those skilled in the art to fully understand the present disclosure, and are not intended to limit the subject matter described in the claims.

In each of the following embodiments, as an example of the entry / exit authentication system according to the present disclosure, an entry / exit management system 5 that authenticates when entering or leaving a conference room inside a facility or the like will be illustrated and described. ..

(Embodiment 1)
FIG. 1 is a diagram showing the inside of a facility 200 in which the entrance / exit management system 5 according to the first embodiment is installed. Facility 200 is, for example, a public facility in which many conference rooms are prepared. Inside the facility 200, a wide passage 130 through which many people pass is provided. A plurality of windows 131 are arranged on one side wall of the passage 130. The other side wall of the aisle 130 is provided with the front of the conference rooms 150, 160, 170. Doors 151, 161, 171 leading to the room are installed in front of each of the conference rooms 150, 160, 170, respectively. Although FIG. 1 shows a plurality of (for example, three) conference rooms, only one conference room may be used.

Each door 151, 161, 171 is provided with an electric lock 158, 168, 178 that can be switched to a locked (locked) or unlocked (unlocked) state, respectively. When the electric lock is switched to the locked state, the door is closed and locked. When the electric lock is switched to the unlocked state, the door is opened and unlocked.

Each door 151, 161 and 171 is provided with a card reader 155, 165, 175, respectively. The card readers 155, 165, and 175 perform non-contact communication with the card cd possessed by the person hm, and read the card information (for example, ID information which is identification information) recorded on the card cd. A face recognition camera 10 is installed on the ceiling or pillar of the passage 130 in front of the doors 151 and 161. Further, a face recognition camera 10 is installed on the upper part of the door 171.

A monitoring room 180 is provided behind the conference rooms 150, 160, 170 in the facility 200. Inside the monitoring room 180, a face authentication server 30, an entrance / exit management server 50, and a PC (Personal Computer) 70 are placed.

FIG. 2 is a diagram showing a configuration example of the entry / exit management system 5. The management of entry and exit in the three conference rooms 150, 160, and 170 is the same. Here, the conference room 160 will be mainly described, but since the installation location of the face recognition camera 10 differs depending on the conference room, other conference rooms 150 and 170 will also be described as needed.

The entry / exit management system 5 includes a face recognition camera 10, a face recognition server 30, an entry / exit management server 50, and a PC 70, which are connected via the network NW. As an example, the face recognition camera 10 is installed on the side wall of the pillar 130z erected in the passage 130. The face recognition camera 10 captures an image including the face of a person hm passing through the passage 130 as a subject. In the first embodiment, the face recognition camera 10 is a constant image pickup camera that constantly captures an image and repeatedly transmits the captured image to the PC 70 and the face recognition server 30. The face recognition camera may be a sensitive camera with a human sensor that captures images only when a person enters the face detection area. Further, although the face recognition camera is a fixed camera here, it may be a PTZ (Pan Tilt Zoom) camera capable of pan / tilt zoom processing.

The face recognition server 30 detects face information (face feature amount, face direction, line-of-sight direction, etc.) of a person hm passing through the passage 130 based on the captured image transmitted from the face recognition camera 10. The entry / exit management server 50 authenticates a person heading for the door 161 and controls the opening / closing of the door 161. The PC 70 sets the door area and the face detection area.

The entry / exit management system 5 also includes a card reader 165, a card cd, and an electric lock 168. The card reader 165 and the electric lock 168 are connected to the entry / exit management server 50. The card reader 165 performs non-contact communication with a card cd possessed by a person. The electric lock 168 automatically opens and closes the door 161 and locks or unlocks the door 161.

FIG. 3 is a diagram showing a hardware configuration example of the face authentication server 30 and the entry / exit management server 50. The face authentication server 30 includes a face detection unit 31, a face direction / line-of-sight detection unit 32, a face information storage unit 33, a face matching unit 34, and a communication unit 35.

The face detection unit 31 detects the face image of a person reflected in the image captured by the face recognition camera 10. The face detection unit 31 is included in the captured image (that is, the current live image captured in real time), and is the position of each feature point such as the eyes, eyebrows, nose, and mouth in the face, and the distance between the feature points. It is judged as a face image based on the above.

The face direction / line-of-sight detection unit 32 detects the face direction and the line-of-sight direction of the detected person. Various methods for detecting the orientation of the face have been conventionally proposed. As an example, the face direction / line-of-sight detection unit 32 rotates based on the position coordinates of feature points such as eyes, nose, and mouth when the face direction (face direction) is estimated, for example, when the face is facing the front. Calculate the matrix, compare the position coordinates after the rotation matrix with the position coordinates of the feature points such as eyes, nose, and mouth included in the face image, and if they are close, the angle (pitch) of the face corresponding to the rotation matrix. Determine the angle, roll angle, yaw angle). Further, as another example, when the face direction / line-of-sight detection unit 32 estimates the direction of the face, the face image is subjected to skin color detection processing, and the center of gravity of the skin color region of the face image facing the front and a predetermined value are obtained. The amount of deviation from the center of gravity of the skin color region of the face image facing the direction is calculated, and the face direction is determined from this amount of deviation.

Further, the direction of the line of sight is determined by using the technique of line-of-sight detection. The line-of-sight detection is performed by detecting the movement of the eyes. The movement of the eyes is specified by the position of the moving part (iris, etc.) of the eye with respect to the position (reference point) of the part where the eye does not move (inner corner, outer corner, etc.). For example, if the iris of the right eye is far from the inner corner of the eye and closer to the outer corner of the eye, it is determined that the person is looking at the right side. Similarly, if the iris of the left eye is far from the inner corner of the eye and closer to the outer corner of the eye, it is determined that the person is looking at the left side. Further, the movement of the eye may be detected from the positional relationship between the corneal reflex and the pupil with the corneal reflex obtained by irradiating the eye with infrared rays as a reference point.

The face information storage unit 33 stores face information of a person, for example, face information of an employee working at a facility 200. The facial information includes a facial image and facial features. The face image includes a front face image and left and right face images of the same person. The facial features include the positional relationship between the eyes, nose, mouth, etc., the presence / absence of glasses, the presence / absence of hair, and the like. Further, the face information storage unit 33 stores the face information of the person when the direction of the face is the direction of the door and the line-of-sight information of the person when the direction of the line of sight is the direction of the door.

The face direction / line-of-sight detection unit 32 determines whether or not the direction of the face detected by the face detection unit 31 is the direction of the door 161 based on the face information stored in the face information storage unit 33. Further, the face direction / line-of-sight detection unit 32 determines whether or not the direction of the line-of-sight detected by the face detection unit 31 is the direction of the door 161 based on the line-of-sight information stored in the face information storage unit 33. .. If both the face direction and the line-of-sight direction are aligned in the direction of the door, the face direction / line-of-sight detection unit 32 may determine that the face direction and the line-of-sight direction are aligned with each other in the direction of the door, or one of them may be in the direction of the door. If, it may be determined that the door is facing the direction of the door. Here, when either one is in the direction of the door, the face direction / line-of-sight detection unit 32 determines that the direction is in the direction of the door.

Further, the face orientation is not limited to the case where the face orientation of the person is detected once, and may be detected for the first time when the orientation of the face of the person is detected more than a predetermined number of times. Further, it may be detected for the first time when the orientation of the face is continuously detected for a predetermined time or longer. That is, when the degree of face orientation is large, the face orientation may be detected. As a result, it is possible to prevent a misjudgment that a person glances at the door, that is, a so-called glance is regarded as an intention to enter or leave. The same applies to the line of sight.

When the direction of the face detected by the face detection unit 31 is the direction of the door 161 or the direction of the line of sight detected by the face detection unit 31 is the direction of the door 161, the face matching unit 34 detects the face. The face information of the face detected by the unit 31 is collated with the face information stored in the face information storage unit 33. For example, the face detection unit 31 is included in the captured image (that is, the current live image captured in real time), and the position of each feature point such as the eyes, eyebrows, nose, and mouth in the face, and the distance between the feature points. It is determined that the face image is based on the distance and the like, and the part of the face image is cut out from the captured image. The face matching unit 34 performs template matching for collating the cut out face image with the face image (template) of each person registered in the face information storage unit 33. The face matching unit 34 calculates the degree of similarity by comparing the feature amount such as the position of each feature point in the cut out face image and the distance between the feature points with the corresponding feature amount of the registered face image. If the degree of similarity exceeds the threshold, face recognition is performed to determine that the face images match.

The face matching unit 34 may perform face recognition using artificial intelligence (AI). In this case, the processor constituting the face matching unit 34 stores the face image of each person as teacher data. Using the above-mentioned teacher data, the processor performs machine learning such as deep learning using artificial intelligence, updates the number of neurons in the middle layer of the neural network that forms the learning model, and the weighting coefficient, and creates a trained model. To construct. The processor may use this trained model to input an image captured by the face recognition camera, authenticate the face image of the person reflected in the captured image, and output the authentication result.

When the face images match as a result of face matching, the face matching unit 34 transmits a permission signal to the electric lock control unit 51 of the entry / exit management server 50 via the communication unit 35 and the network NW. Upon receiving the permission signal from the face matching unit 34, the electric lock control unit 51 unlocks the electric lock 168 and automatically opens the door 161. If the face matching unit 34 does not match the face images as a result of the face matching, the face matching unit 34 does not transmit the permission signal. Therefore, the electric lock 168 is not unlocked and the door 161 remains closed.

The communication unit 35 is connected to the network NW by wire or wirelessly, and communicates with the entry / exit management server 50, the face recognition camera 10 and the PC 70 via the network NW. The communication unit 35 may be directly connected to the entry / exit management server 50, the face authentication camera 10 or the PC 70 without going through the network NW, and may communicate with them.

The entry / exit management server 50 includes an electric lock control unit 51, a communication unit 52, a card reader communication unit 53, an ID position detection unit 54, an ID information storage unit 55, an ID authentication unit 56, and an ID information setting unit 57.

The electric lock control unit 51 drives the electric lock 168 and controls the opening and closing of the door 161 by the electric lock 168. The communication unit 52 is connected to the network NW by wire or wirelessly and communicates with the face recognition server 30. The communication unit 52 may be directly connected to the face authentication server 30 and communicate with the face authentication server 30 without going through the network NW.

The card reader communication unit 53 performs non-contact communication with the card reader 165 and acquires the card information read by the card reader 165. The card reader 165 performs non-contact communication with a card cd possessed by a person by using a short-range wireless communication method. As the short-range wireless communication method, for example, Bluetooth (registered trademark), Felica (registered trademark), or the like may be used. The card reader 165 is not limited to the non-contact communication method, and may be a method of reading the information recorded on the card cd in a state of being in contact with the card cd (that is, contact communication).

The ID position detection unit 54 detects the ID information included in the card information acquired by the card reader communication unit 53. The ID information includes information for identifying a person registered on the card cd. The ID position detection unit 54 detects not only the ID information but also the position information of the card reader 165 (for example, the identification information of the door on which the card reader 165 is arranged) from the card information acquired by the card reader communication unit 53. You may.

The ID information storage unit 55 stores ID information of many people and information about each person in association with each other. For example, as a person, an employee in the facility 200 and the like can be mentioned.

The ID authentication unit 56 collates the ID information detected by the ID position detection unit 54 with the ID information stored in the ID information storage unit 55, and as a result of the collation, the ID authentication is approved (OK). ), The permission signal is output to the electric lock control unit 51. Upon receiving the permission signal from the ID authentication unit 56, the electric lock control unit 51 unlocks the electric lock 168 and automatically opens the door 161. If the ID authentication is rejected (NG) as a result of the collation, the ID authentication unit 56 does not transmit the permission signal. Therefore, the electric lock 168 is not unlocked and the door 161 remains closed.

The ID information setting unit 57 accepts an operation by an operator or the like, sets the ID information of the person and the information about the person, and registers the information in the ID information storage unit 55.

Here, two operations are possible, one is to authenticate the person by face verification and the other is to confirm the person by ID authentication. When authenticating a person by face matching, it takes time to process, but the authentication accuracy of whether or not the person is the person is high, and security is improved. On the other hand, when confirming a person by ID authentication, it is assumed that the person of the information recorded on the card is different from the person holding the card, but the processing time is shortened. The first embodiment shows a case where a person is authenticated by face matching, and the second embodiment shows a case where a person is confirmed by ID authentication. In addition, the case of authenticating a person by face verification and the case of confirming a person by ID authentication may be used together. As a result, security can be further improved.

FIG. 4 is a diagram showing a setting screen GM1 in which the door region DA1 and the face detection area SA1 are set in the captured image GZ1 from the side of the door 161. The setting screen GM1 is displayed on the PC 70 operated by an operator or the like. On the setting screen GM1, the captured image GZ1, the setting button bt1, and the delete button bt2 are displayed. In the case of the conference room 160, the captured image GZ1 is a live image captured by the face recognition camera 10 installed on the pillar 130z of the passage 130 and viewed from the side in front of the door including the door 161. The door 161 is a sliding door that slides left and right to open and close. A card reader 165 is attached to the side of the door 161. The captured image GZ1 shows a person hm approaching the door 161.

The setting button bt1 is pressed by an operator or the like when starting the setting of the door area DA1 and the face detection area SA1 and when saving the set door area DA1 and the face detection area SA1. The delete button bt2 is pressed by an operator or the like when erasing the door area DA1 and the face detection area SA1 that have already been set, or the door area DA1 and the face detection area SA1 that are being set.

The operator sets the door area DA1 so as to surround the door 161 included in the captured image GZ1 by using the cursor Ks operated by the mouse. On the setting screen GM1, the set door area DA1 is drawn with a red dotted line frame. When the door area DA1 is set, the PC 70 automatically sets the face detection area SA1 within a range of a predetermined distance from the door 161, for example, within a range of 3 m from the front of the door 161. The operator may edit the automatically set face detection area SA1 with the cursor Ks. On the setting screen GM1, the set face detection area SA1 is drawn with a blue dotted line frame.

Here, the operator sets the door area DA1, but the door area DA1 may be set so that the PC 70 detects the door image included in the captured image GZ1 by image processing and surrounds the detected door image. When the door area DA1 is set, the PC 70 automatically sets a predetermined range as the face detection area for the door position, but the operator may set the face detection area SA1. The face detection area may detect a person and may be set in a predetermined range centered on the height of the person. For example, an area of 1/2 or more of the height of the person may be set as the face detection area. The face detection area is set to a rectangular area, but may be set to any shape. The face detection area is not limited to one, and a plurality of face detection areas may be set.

FIG. 5 is a diagram showing a setting screen in which the door area DA2 and the face detection area SA2 are set in the captured image GZ2 from the front direction of the door 161. In the case of the conference room 170, the captured image GZ2 is a live image captured by the face recognition camera 10 installed above the door 161 and viewed from the front of the door. The captured image GZ2 shows a person hm approaching the door 161 while running. The door 161 is not shown in the captured image GZ2. In this case, the operator uses the cursor Ks operated by the mouse to specify the direction in which the door is installed. For example, specify by dragging or double-clicking the mouse at the screen position in the direction in which the door is installed. When the door area DA2 is set, the door area DA2 is drawn with a red dotted line frame on the setting screen GM2. When the door area DA2 is set, the PC 70 automatically sets the face detection area SA2 within a range of a predetermined distance from the door 161, for example, within a range of 3 m from the front of the door 161. The operator may edit the automatically set face detection area SA2 with the cursor Ks. On the setting screen GM2, the set face detection area SA2 is drawn with a blue dotted line frame.

As in the case of the setting screen GM1, when the door area DA2 is set, the PC 70 automatically sets a predetermined range as the face detection area for the door position, but the operator sets the face detection area SA2. You may. The face detection area may detect a person and may be set in a predetermined range centered on the height of the person. For example, an area of 1/2 or more of the height of the person may be set as the face detection area. The face detection area is set to a rectangular area, but may be set to any shape such as an ellipse or a polygon. The face detection area is not limited to one, and a plurality of face detection areas may be set.

Next, the operation procedure of the entry / exit management system 5 according to the first embodiment will be described with reference to FIG.

FIG. 6 is a sequence diagram showing a setting operation in the entry / exit management system 5.

In FIG. 6, the face recognition camera 10 constantly images the vicinity of the door and transmits the captured image to the PC 70 (T1). The PC 70 receives the image captured by the face recognition camera 10 and displays the received image on the screen. The PC 70 accepts an operation by the operator for setting the door area for the captured image displayed on the screen (T2).

The PC 70 automatically sets the face detection area based on the door area set by the operator. The PC 70 transmits information including the door area and the face detection area set in association with the door area to the face authentication server 30 (T3). When the communication unit 35 of the face authentication server 30 receives the information of the door area and the face detection area from the PC 70, the communication unit 35 stores the information in the face information storage unit 33.

FIG. 7 is a sequence diagram showing the operation of the entry / exit management system 5 during operation. The face recognition camera 10 captures a person hm passing through the passage 130 (T11). The face recognition camera 10 transmits the captured image to the face detection unit 31 of the face recognition server 30 (T12). The face detection unit 31 detects the face image included in the captured image and notifies the face direction / line-of-sight detection unit 32 of this face information (including the face image) (T13). The face direction / line-of-sight detection unit 32 determines the direction of the face (face direction) and the direction of the line of sight based on the face information. That is, it is determined whether the face direction is the door direction or the line-of-sight direction is the door direction. When the face direction or the line-of-sight direction is the door direction, the face direction / line-of-sight detection unit 32 transmits the determination result of the face direction or the line-of-sight direction to the face matching unit 34 (T14).

The face collation unit 34 collates the face image detected by the face detection unit 31 with the face information (face image, facial feature amount, etc.) of the person stored in the face information storage unit 33 (T15). The processes of steps T12 to T15 are repeated while the person is in the face detection area.

When the collation is OK, the face collation unit 34 transmits a permission signal (door opening control information) to the entry / exit management server 50 via the communication unit 35 and the network NW (T16).

The communication unit 52 of the entry / exit management server 50 receives the door opening control information from the face authentication server 30. The communication unit 52 transmits the door opening control information to the electric lock control unit 51. The electric lock control unit 51 unlocks the electric lock 168 and opens the door 161 (T17). As a result, the door 161 opens automatically.

FIG. 8 is a diagram showing an operating state for a plurality of persons passing in front of the door 161 when the angle of view of the face recognition camera 10 is set to a wide angle. In FIG. 8, the positions of the door 161 and the face recognition camera 10 are schematically drawn for the sake of clarity. The same applies to FIGS. 9 and 10. The face recognition camera 10 installed above the passage 130 captures images of three people hm4 to hm6 passing in front of the door 161 and the door 161 so as to be included in the angle of view. In this case, the first person hm4 among the three people hm4 to hm6 faces the door 161 while the remaining two people hm5 and hm6 point in a different direction than the door 161. It is suitable. The entrance / exit management system 5 operates so as to automatically open the door 161 assuming that the leading person hm4 is facing the door 161. Therefore, the entrance / exit management system 5 can control the unlocking of the door 161 so that only the leading person hm4 can enter and leave the room, and prohibits the unlocking of the door 161 so that the person hm5 and hm6 cannot enter and leave the room. ..

FIG. 9 is a diagram showing an operating state for a plurality of persons passing in front of the door 161 when the angle of view of the face recognition camera 10 is set to a narrow angle. The face recognition camera 10 captures only the first person hm4 (the size of one person hm) among the three people hm4 to hm6 passing in front of the door 161. The leading person hm4 is facing the door 161 while the remaining two people hm5 and hm6 are facing a different direction than the door 161. In this case, the entrance / exit management system 5 operates so as to automatically open the door 161 assuming that the leading person hm4 is facing the door 161. In the first embodiment, regardless of whether the angle of view of the face recognition camera 10 is wide-angle or narrow-angle, the time during which the door 161 is controlled to be unlockable by the entrance / exit management system 5 is the time during which one person can pass. Is. When the leading person hm4 passes, the door 161 closes. That is, the entrance / exit management system 5 closes the door 161 until the next person hm5 is imaged by the face authentication camera 10 and it is determined that the face is facing the door 161. Therefore, it is possible to prevent the actions of the persons hm5 and hm6 who try to enter the room with their tailgating without turning their faces to the door. When it is determined that the persons hm5 and hm6 are facing the door 161 respectively, the entrance / exit management system 5 opens the door 161 at each of them.

FIG. 10 is a diagram showing an operating state for a plurality of persons hm4 to hm6 passing in front of the door 161 when two face recognition cameras 10 are installed. In this case, the rear face recognition camera 10 can capture the person hm from behind. For example, assume that there is a person hm7 who is an ineligible person behind the person hm4 at the head. In this case, the rear face recognition camera 10 can take an image of the person hm7 behind. Therefore, it is also possible to find and track the person hm7 who is going to enter and leave with tailgating.

As described above, in the entry / exit management system 5 according to the first embodiment, the face authentication server 30 (an example of the face authentication device) and the entry / exit management server 50 (an example of the entry / exit management device) are communicably connected to each other. The face recognition server 30 is based on an image captured by the face recognition camera 10 that captures a person hm (an example of a person) located near the door 161 (an example of an entrance) of the conference room 160 (an example of a managed area). It detects whether or not the person hm is facing the door 161 and sends the detection result to the entry / exit management server 50. When the entry / exit management server 50 detects that the person hm is facing the door 161 based on the detection result sent from the face authentication server 30, the person hm enters or exits the conference room 160. The electric lock 168 of the door 161 is locked or unlocked (an example of the open / closed state) according to the determination result of whether or not the person is authorized.

As a result, the entrance / exit management system 5 can effectively suppress the entry / exit of a person who does not intend to enter / leave the room, so that it is possible to eliminate the intrusion of outsiders and improve the security. Further, according to the entry / exit management system 5, since a person is authenticated by face matching, there is a high possibility that the person (for example, an employee) is willing to enter / leave the room, and high security can be ensured.

Further, the face authentication server 30 detects the face image of the person hm from the images captured by the face authentication camera 10, and detects whether or not the detected face image of the person hm faces the door 161. As a result, according to the entrance / exit management system 5, it is possible to appropriately select a person who intends to enter / leave the room from the passersby. Therefore, according to the entrance / exit management system 5, the chance of unnecessarily unlocking and opening the door is reduced. Further, according to the entry / exit management system 5, since the gesture of turning to the face is a natural action performed by a person who has an intention to enter / leave the room, it is not necessary to force the person to show the intention to enter / leave the room.

Further, the face authentication server 30 detects the face image of the person hm from the captured image of the face authentication camera 10, and detects whether or not the line of sight in the detected face image is directed to the door 161. As a result, according to the entrance / exit management system 5, it is possible to respond appropriately without repelling a person who makes a gesture of turning only his eyes toward the door without turning his face toward the door. Further, according to the entry / exit management system 5, it is possible to easily communicate the intention without making a large face-turning action.

Further, the face authentication server 30 starts detecting the face image of the person hm in response to the detection that the person hm has entered the face detection area SA1 set in the captured image of the face authentication camera 10. As a result, according to the entry / exit management system 5, a person outside the face detection area can be excluded from the target by setting a face detection area in an area where it is assumed that a person who intends to enter / exit turns his / her face. can. Therefore, according to the entry / exit management system 5, the processing load can be reduced. Further, according to the entry / exit management system 5, the face detection is performed at the timing when the person hm enters the face detection area, so that the person who intends to enter / leave the room can be selected with high accuracy.

Further, the face detection area SA1 is set within a predetermined distance from the door 161. Thereby, according to the entrance / exit management system 5, the face detection area can be appropriately set for, for example, the door 161.

Further, the face detection area SA1 has the size of one person hm that appears in the image captured by the face recognition camera 10. As a result, according to the entry / exit management system 5, individual persons can be selected even when there is so-called tailgating, in which friends pass by. Therefore, according to the entry / exit management system 5, it is possible to eliminate a malicious person who tries to enter / leave by pretending to be a tailgating.

Further, the face recognition camera 10 is arranged at a position where both the door 161 and the person hm approaching the door 161 are included in the angle of view. As a result, according to the entrance / exit management system 5, the face recognition camera 10 can be arranged on the side of the door 161. In addition, the face recognition camera 10 can reliably capture a person who intends to enter or leave the room. The operator can individually confirm whether or not the person heading for the door is a person who intends to enter or leave by looking at the captured image.

Further, the face recognition camera 10 is arranged at a position where the person hm approaching the door 161 is included in the angle of view. As a result, according to the entrance / exit management system 5, the face recognition camera 10 can be attached to the door 161. Further, according to the entry / exit management system 5, the face of a person reflected in the captured image can be captured from the front. Therefore, according to the entry / exit management system 5, false detection can be reduced and the accuracy of face detection can be improved.

(Embodiment 2)
In the first embodiment, a case where a person is authenticated by face matching is shown. The second embodiment shows a case where a person is confirmed by ID authentication. The entry / exit management system according to the second embodiment has substantially the same configuration as that of the first embodiment. The same components as those in the first embodiment are used with the same reference numerals, and the description thereof will be omitted.

FIG. 11 is a sequence diagram showing an operation during operation of the entry / exit management system 5 of the second embodiment. Since the operations in the procedures T12 to T13 are the same as those in the first embodiment, the description of the duplicated contents will be simplified or omitted, and different contents will be described.

When the face direction / line-of-sight direction is the direction of the door 161, the face direction / line-of-sight detection unit 32 transmits the face direction / line-of-sight information to the entry / exit management server 50 via the communication unit 35 and the network NW ( T21). In this case, the face matching unit 34 passes the face direction / line-of-sight information to the communication unit 35 as it is. The communication unit 52 of the entry / exit management server 50 notifies the ID authentication unit 56 of the face direction / line-of-sight information.

The ID authentication unit 56 collates the ID information detected by the ID position detection unit 54 with the ID information stored in the ID information storage unit 55, and performs ID authentication (T22). If the ID authentication is OK as a result of the collation, the ID authentication unit 56 outputs a permission signal (door opening control information) to the electric lock control unit 51 (T23). Upon receiving the door opening control information from the ID authentication unit 56, the electric lock control unit 51 unlocks the electric lock 168 and opens the door 161 (T24). As a result, the door 161 is automatically opened.

As described above, in the entry / exit management system according to the second embodiment, the person can be quickly confirmed by ID authentication by reading the ID information (an example of the identification information) stored in the card cd possessed by the person. Further, according to the entry / exit management system 5 according to the second embodiment, erroneous detection is unlikely to occur.

Although various embodiments have been described above with reference to the drawings, it goes without saying that the present disclosure is not limited to such examples. It is clear that a person skilled in the art can come up with various modifications, modifications, substitutions, additions, deletions, and even examples within the scope of the claims. It is understood that it naturally belongs to the technical scope of the present disclosure. Further, each component in the various embodiments described above may be arbitrarily combined as long as the gist of the invention is not deviated.

For example, in the above-described first and second embodiments, the opening and closing of the door for entering and exiting the room (meeting room) in the facility is shown, but for entering and exiting the front entrance of the facility or a place such as a plaza or a park. The present disclosure is similarly applicable to the opening and closing of the gate of.

This disclosure is useful as an entry / exit authentication system and an entry / exit authentication method for effectively suppressing entry / exit of a person who does not intend to enter / leave the room and improving security.

5 Access control system 10 Face recognition camera 30 Face recognition server 31 Face detection unit 32 Face direction / line-of-sight detection unit 33 Face information storage unit 34 Face verification unit 35 Communication unit 50 Access control unit 50 Electric lock control unit 52 Communication unit 53 Card reader communication unit 54 ID position detection unit 55 ID information storage unit 56 ID authentication unit 57 ID information setting unit 70 PC
161 Door 165 Card Reader 168 Electric Lock

Claims (9)

It is an entry / exit authentication system in which a computer, a face recognition device, and an entrance / exit management device are connected so as to be communicable.
The computer
The setting screen of the entrance and the face detection area of the person including the image captured by the camera that captures the image of the person located near the entrance of the managed area is displayed, and the detection based on the image captured by the camera or the user operation is performed. The entrance and the face detection area are set and sent to the face recognition device.
The face recognition device is
Based on the setting result of the entrance and the face detection area from the computer and the image captured by the camera, it is detected whether or not the person is facing the entrance, and the detection result is transmitted to the entrance / exit management device. Send,
The entry / exit management device is
When it is detected that the person is facing the entrance based on the detection result sent from the face recognition device, the person is a person who is permitted to enter or leave the controlled area. The open / closed state of the entrance is controlled according to the determination result of whether or not.
Entry / exit authentication system. The face recognition device is
A face image of the person is detected from the images captured by the camera, and it is detected whether or not the detected face image of the person faces the entrance.
The entry / exit authentication system according to claim 1. The face recognition device is
A face image of the person is detected from the captured image of the camera, and it is detected whether or not the line of sight in the detected face image is directed to the entrance.
The entry / exit authentication system according to claim 1. The face recognition device is
In response to the detection that the person has entered the face detection area set in the image captured by the camera, the detection of the face image of the person is started.
The entry / exit authentication system according to claim 2 or 3. The face detection area is set within a predetermined distance from the entrance.
The entry / exit authentication system according to claim 4. The face detection area further has the size of one person appearing in the image captured by the camera.
The entry / exit authentication system according to claim 5. The camera is placed at a position where both the entrance and the person approaching the entrance are included in the angle of view.
The entry / exit authentication system according to claim 1. The camera is arranged at a position within the angle of view of a person approaching the entrance.
The entry / exit authentication system according to claim 1. It is an entry / exit authentication method using an entry / exit authentication system in which a computer, a face recognition device, and an entrance / exit management device are connected so as to be communicable.
The computer
The setting screen of the entrance and the face detection area of the person including the image captured by the camera that captures the image of the person located near the entrance of the managed area is displayed, and the detection based on the image captured by the camera or the user operation is performed. The entrance and the face detection area are set and sent to the face recognition device.
The face recognition device is
Based on the setting result of the entrance and the face detection area from the computer and the image captured by the camera, it is detected whether or not the person is facing the entrance, and the detection result is transmitted to the entrance / exit management device. Send,
The entry / exit management device is
When it is detected that the person is facing the entrance based on the detection result sent from the face recognition device, the person is a person who is permitted to enter or leave the controlled area. The open / closed state of the entrance is controlled according to the determination result of whether or not.
Entry / exit authentication method.

Images