JP7298952B1 - Image determination device, traffic management device, and image determination method - Google Patents

Abstract

Kind Code: A1 The present invention determines the presence or absence of spoofing using a minimum number of cameras. A plurality of cameras (1A, 1B) that capture predetermined points in a detection target area at different angles of view, and an image acquisition function unit that acquires a target image included in image data captured by these cameras (1A, 1B). 2, a correlation detection function unit 3 that detects the positional correlation in the detection target area of the target images respectively acquired by the image acquisition function unit 2 from the image data captured by the one camera 1A and the other camera 1B; A spoofing judgment function unit 4 for judging whether or not the target image in the image data of the first camera 1A and the other camera 1B is a spoofed image from the target image having the correlation detected by the correlation detection function unit 3. and [Selection drawing] Fig. 1

Description

The present invention relates to an image determination device, a traffic management device, and an image determination method.

Japanese Patent Application Laid-Open No. 2002-200310 discloses a technology related to spoofing detection in face authentication. In Patent Document 1, two types of images are used: an image obtained from a first camera and an image obtained from a second camera that captures images from a direction difficult to obtain from the first camera. doing the analysis.
In addition, in Patent Documents 2 and 3 related to the present invention, by converting feature points from images of a plurality of cameras into images in a three-dimensional space, image analysis with higher accuracy is performed. We are trying to prevent spoofing by companions.

JP 2008-108243 A WO2021/060256 JP-A-2004-192378

However, the image analysis techniques described in Patent Literatures 1 to 3 require processing for generating a three-dimensional image from images obtained from a plurality of cameras, so it takes time to generate the image. It can be suitably adopted for active authentication such as door unlocking in building security management, but it is practical for non-active authentication such as walk-through authentication and where high-speed processing that does not hinder the flow of people is required. The problem is that it is difficult to convert
In addition to the above Patent Documents 1 to 3, as a method of preventing spoofing by capturing an image of a display device or the like into a camera, a camera that is easily recognized by the target person (the target person is likely to point the fake image at the camera) and a recognition method. There is also a method that uses images from a camera paired with a camera that is difficult to be stolen (a hidden camera that makes it difficult for the target person to point a fake image at the camera), and uses the results of face recognition from the images of both cameras to determine spoofing. However, in the case of a two-way facility with two lanes of one-way walk-through gates, a total of two pairs of cameras are required, one pair for each lane, and authentication processing is required for the two pairs of camera images. As the amount of processing increases due to the processing, there is a problem that the processing time increases and the preparation of hardware with higher performance becomes necessary.

An object of the present invention is to determine spoofing using a limited number of cameras.

In order to solve the above problems, the present invention proposes the following means.
An image determination apparatus according to a first aspect of the present invention is characterized by the following configuration.
This image determination device includes a plurality of cameras that capture predetermined points in a detection target area at different angles of view, an image detection function unit that detects a target image included in image data captured by these cameras, and a single camera. A correlation detection function unit for detecting a positional correlation in the detection target area of the target image detected by the image detection function unit from image data captured by a camera and another camera, and a correlation detected by the correlation detection function unit. and a spoofing determination function unit that determines whether the target image in the image data of the one camera and the other camera is a spoofed image, from the target image having the

An image determination method according to a second aspect of the present invention is characterized by the following configuration.
This image determination method includes a step of photographing a predetermined point in the detection target region L with a plurality of cameras having different angles of view, a target image detection step of detecting a target image included in the captured image data, and a detecting the positional correlation in the detection target area of the target images respectively detected from the image data captured by the camera and the other camera; and determining whether or not the target image in the image data is a spoofed image.

According to the present invention, spoofing can be determined by images obtained from at least two cameras.

1 is a block diagram of a minimum configuration example of an image determination apparatus according to the present invention; FIG. 1 is a flowchart of a minimum configuration example of an image determination method according to the present invention; 1 is an overall configuration diagram of an embodiment of the present invention; FIG. 1 is a block diagram of one embodiment; FIG. It is a block diagram at the time of judgment preparation of one embodiment. FIG. 10 is a block diagram during correlation detection according to an embodiment; It is a block diagram at the time of standby of one embodiment. It is a block diagram at the time of passerby detection of one embodiment. It is a block diagram at the time of data correspondence of one embodiment. It is a block diagram at the time of determination of one embodiment. FIG. 4 is an explanatory diagram of rectangle association according to one embodiment; FIG. 4 is an explanatory diagram of correlation determination according to one embodiment;

A configuration of an image determination apparatus according to the minimum configuration of the present invention will be described with reference to FIG.
This device includes a plurality of cameras 1A and 1B that capture predetermined points in a detection target area at different angles of view, and an image detection function unit that detects a target image included in image data captured by these cameras 1A and 1B. 2, a correlation detection function unit 3 that detects the positional correlation in the detection target area of the target images respectively detected by the image detection function unit 2 from the image data captured by the one camera 1A and the other camera 1B; A spoofing judgment function unit 4 for judging whether or not the target image in the image data of the first camera 1A and the other camera 1B is a spoofed image from the target image having the correlation detected by the correlation detection function unit 3. and

Then, according to the determination device according to the present invention configured as described above, images photographed by one camera and another camera are compared in a common target area, and images having a correlation are compared with each other, and the comparison result is The presence or absence of spoofing can be determined from the

An image determination method according to the minimum configuration example of the present invention will be described with reference to FIG.
This image determination method includes a step SP1 of photographing a predetermined point in a detection target area with a plurality of cameras having different angles of view, a target image detection step SP2 of detecting a target image included in the photographed image data, a step SP3 of detecting positional correlation in the detection target area of target images respectively detected from image data captured by a camera and another camera; and a step SP4 of determining whether or not the target image in the image data is a spoofed image.

Then, according to the determination method according to the present invention configured as described above, images photographed by one camera and another camera are compared in a common target area, and images having correlation are compared with each other, and the comparison result is The presence or absence of spoofing can be determined from the

A configuration according to an embodiment of the present invention embodying FIGS. 1 and 2 will be described with reference to FIGS. 3 to 12. FIG.
FIG. 3 shows the overall configuration of the passage determination device of one embodiment.
Face authentication device 100 receives input of image data from camera 200 and camera 201 that shoot in opposite directions. The face authentication device 100 is also connected to a door-shaped gate 300 that opens and closes a traffic lane formed between the side walls on both sides. It outputs signals for operating and/or locking and unlocking (locking and unlocking) the gate 300 .
The walk-through gate controls passage in both left and right directions in the drawing, and one of the cameras 200 and 201 always photographs the point of the gate 300 in the direction of entry, and the other photographs in the direction of exit. are doing. That is, the cameras 200 and 201 photograph users in the entry lane and the exit lane in the direction intersecting the gate 300, respectively, in order to authenticate persons in both directions of entry and exit.
When the authentication device 100 detects the human body and/or face of the subject about to pass through the gate 300, the authentication device 100 combines the detection results of both cameras 200 and 201 with the gate 300 calculated and associated in advance. and the coordinate information of the position corresponding to . The details of the determination processing will be described later.

FIG. 4 is a block diagram showing the configuration of the face authentication device 100. As shown in FIG.
Face authentication software 110 and 120 installed in this face authentication device 100 receives image data supplied from the cameras 200 and 201 by image acquisition function units 111 and 121, respectively, and detects face/human body detection function units 112 and 122. , positional correlation detection function units 113 and 123 and detection information synthesis function units 115 and 125 .
The face authentication software 110, 120 detects the positional correlation detection function units 113, 123 with reference to the reference data of the correlation in the comparison image databases 114, 124. A correlation value is calculated by the comparison, supplied to the detection information synthesizing function units 115 and 125, and comprehensively evaluated together with judgment of the image data. (Passing through the gate) is determined, and based on these determination results, the comprehensive determination function units 117 and 127 determine whether passing through is permitted. Then, based on the judgment results of these two comprehensive judgment function units 117 and 127, for example, on the condition that both of the comprehensive judgment function units 117 and 127 have issued judgment results that impersonation is not possible (passage is allowed), the The lock determination function unit 130 determines that unlocking is permitted, unlocks the gate 300, and operates an actuator or the like to open the gate.

The operation of the face authentication device configured as described above will be described together with the steps of the face authentication method.
As shown in FIG. 5, as an initial procedure, the face authentication device 100 performs a process of setting basic data for correlating the positions of the detection results of the two face authentication softwares 110 and 120 .
The face authentication software 110 acquires image data captured by the camera 200 using the image acquisition function unit 111 . The face/human body detection function unit 112 in the face authentication software 110 detects a subject to be used as a reference. , 4 pixels×4 pixels), and information about which rectangle the detected person belongs to is stored in the comparison image database 114 . Similarly, the face authentication software 120 acquires image data captured by the camera 201 from the opposite direction of the camera 200 using the image acquisition function unit 121 . The face/human body detection function unit 122, after detecting a reference subject, decomposes the current image input by the positional correlation detection function unit 123 into rectangles of a predetermined size (for example, 4 pixels×4 pixels), Information about which rectangle the detected person belongs to is stored in the comparison image database 124 .

In this way, in both face recognition software 110 and 120, the image information is decomposed into rectangles, and processing is performed to analyze which rectangle the detected person belongs to, and information about which rectangle the detected person belongs to is saved. After that, as shown in FIG. 6, the positional correlation detection function unit 113 transmits the detection information to the positional correlation detection function unit 123 . As shown in FIG. 11, the positional correlation detection function unit 123 associates each rectangle between the image a of the camera 200 and the image b of the camera 201, and then uses the result of the association as the positional correlation detection function. The image is returned to the functional unit 113 and stored in the comparison image database 124 at the same time. The positional correlation detection function unit 113 stores the received association result in the comparison image database 114 . That is, both comparison image databases 114, 124 store the result of each rectangle association.

The correlation performed by the positional correlation detection function unit 123 is, for example, obtained by acquiring image data captured by the cameras 200 and 201 from mutually opposite directions (directions different by 180 degrees), as shown in FIGS. As shown in and , it is an image with a symmetrical contour shape (silhouette is symmetrical), and it refers to the mutual association of rectangles that have a corresponding positional relationship (exist on the same coordinates) in this symmetrical image. Specifically, as shown in FIG. 11, the outer edge of the gate of one image (a) corresponds to the outer edge of the other image (b), and the inner side of the gate of one image (a) corresponds to the inner edge of the other image (b). The area (the area specified by the predetermined coordinates around the gate) where the image of each part of the passerby's body (at least the head, which is likely to be spoofed on the premise of face authentication) between the gates is captured , and likewise each rectangle is associated with each of the images captured by the left and right cameras 200 and 201 .

After the initial procedure by the above steps is completed, the face authentication device 100 is once put into a standby state as shown in FIG. Wait for the data to be output (wait for the image to change as the passerby is photographed).
When a passerby passes the walk-through gate in this standby state, the camera 201 on the side facing the passerby takes an image, and the image data acquired by the image acquisition function unit 121 is used by the face/human body detection function unit. When 122 detects a face and a person, it sends the result to comprehensive determination function 127 and positional correlation detection function 123, as shown in FIG.

The positional correlation detection function unit 113 acquires the position of the detected person based on the data stored in the comparison image database 124 . On the other hand, when a passerby enters the walk-through gate and the passerby is reflected in the camera 200 behind the passerby, the face/human body detection function detects the face/human body from the image data acquired by the image acquisition function unit 111 . The unit 112 detects a person and transmits the result to the positional correlation detection function unit 113 .

As shown in FIG. 8, the positional correlation detection function unit 113 acquires the position of the detected person based on the data of the comparative image database 114 and transmits the positional correlation detection function unit 123 . As shown in FIG. 9, the other positional correlation detection function unit 123 receives the data from the one positional correlation detection function unit 113, and sends the detected information synthesizing function unit 125 to the one positional correlation function unit 113 and the other position detection function unit 125. As shown in FIG. The result detected by the function unit 123 is transmitted. As shown in FIG. 12, the detection information synthesizing function unit 125 associates, for example, the temporal regions of the passerby's outer gate side in the images (a) and (b) with each other, and Both received data are associated with each other so as to correspond to each other, and are transmitted to the spoofing determination function unit 126 .

As shown in FIG. 10, based on the received data, the spoofing determination function unit 126 determines whether a real person (for example, a The unlocking determination function by determining whether the determination result analyzed for each of the rectangles (a) and (b) in one direction and the opposite direction expressed two-dimensionally is a mismatched face image). It notifies the unit 130 of the determination result as to whether or not it is impersonation. At the same time, it receives the collation result from the comprehensive judgment function unit 127, and based on the information, the unlock judgment function unit 130 judges whether or not the lock can be finally unlocked. The gate 300 is controlled to perform operations such as opening the gate, unlocking the gate, etc. In other cases, the gate is not opened or unlocked, and a warning display or the like is displayed.
In the actual operation of the gate, in addition to the above spoofing determination, the relative relationship of facial features such as contours, eyes, ears, nose, mouth, etc. Based on the determination result of face authentication, which determines the identity by comparing the commonality of features with reference data, the gates are controlled to be opened/closed and locked/unlocked.

As described above, in one embodiment, the passers-by are photographed in the entrance direction and the exit direction, respectively, and the facility for performing authentication based on the photographed images is used as it is, and the images of one camera and the other camera are used. By determining whether the images of the corresponding portions match, it is possible to determine the presence or absence of spoofing without adding special image acquisition equipment for spoofing determination.

As described above, one embodiment of the present invention has been described in detail with reference to the drawings, but the specific configuration is not limited to this embodiment, and design changes and the like are also included within the scope of the present invention.

INDUSTRIAL APPLICABILITY The present invention can be used for an image determination device, a traffic management device, and an image determination method in entrance/exit management and the like.

1A, 1B camera 2 image acquisition function unit 3 correlation detection function unit,
4 spoofing determination function unit 100 face authentication device 110, 120 face authentication software 111, 121 image acquisition function unit 112, 122 face/human body detection function unit 113, 123 position correlation detection function unit 114, 124 comparison image database 115, 125 detection information Synthesis function unit 116, 126 (Spoofing) determination function unit 117, 127 Comprehensive determination function determination unit 130 Unlock determination function unit 200, 201 Camera 300 Gate

Claims (5)

a plurality of cameras that capture predetermined points in a detection target area at different angles of view;
an image acquisition function unit that acquires a target image included in image data captured by these cameras;
a correlation detection function unit that detects the positional correlation in the detection target area of the target images respectively detected by the image acquisition function unit from the image data captured by one camera and the other camera;
a spoofing determination function unit that determines whether the target image in the image data of the one camera and the other camera is a spoofed image from the target image having the correlation detected by the correlation detection function unit. death,
The correlation detection function unit decomposes image data of a subject photographed by the plurality of cameras into target images of a plurality of portions,
The spoofing determination function unit determines whether or not the subject is real by comparing the target image at the point of each decomposed portion.
Image judgment device. the plurality of cameras photographing the predetermined points from opposite directions;
The image determination device according to claim 1. The face/human body detection function unit decomposes the image data into rectangles of a predetermined size and detects which rectangle the target image belongs to,
The correlation detection function unit associates and stores the rectangles of the image of one camera and the image of the other camera.
The image determination device according to claim 1 or 2 . The image determination device according to any one of claims 1 to 3 ,
a gate device that regulates the passage of people in the area photographed by the camera;
an unlocking determination unit that cancels the traffic regulation by the gate device according to the determination result of the image determination device;
A traffic management device with a a step of photographing a predetermined point in the detection target area with a plurality of cameras having different angles of view;
a target image detection step of acquiring a target image included in captured image data;
a step of detecting positional correlation in the detection target area of target images respectively detected from image data captured by one camera and another camera;
determining whether or not the target image in the image data of the one camera and the other camera is a spoofed image from the target image having the detected correlation;
has
The step of detecting the positional correlation includes decomposing image data of the subject photographed by the plurality of cameras into target images of a plurality of portions, respectively;
The step of determining whether or not the image is a spoofed image includes determining whether or not the subject is substantial by comparing the target image at the point of each decomposed portion.
Image judgment method.

Images