Mobile body authentication system

Abstract

To provide a technique capable of reducing a processing load when a moving body is authenticated in a predetermined area.SOLUTION: A moving body authentication system includes a processor capable of communicating with a plurality of cameras present in a predetermined area, and a storage device for storing an authentication image for a moving body and camera information indicating at least the position and direction of each of the plurality of cameras. Ease of authentication of the moving body depends at least on distance between the moving body and the camera, and increases as the distance decreases. When a request to authenticate the moving body is received, the processor acquires position information indicating a current position of the moving body, and selects a target camera whose ease of authentication is equal to or higher than a predetermined level from among the plurality of cameras based on at least the position information and the camera information. Further, the processor performs authentication of the moving body based on comparison between the image by the target camera and the authentication image.SELECTED DRAWING: Figure 4

Description

[Technical field]

[0001] The present disclosure relates to techniques for authenticating mobile objects. [Background technology]

[0002] Patent Document 1 discloses a monitoring system. The monitoring system detects the person being watched based on the face recognition data of the person being watched and the face recognition information by a camera, and records a certain amount of video. The monitoring system then transmits URL information for accessing the video to the email address of the communication terminal of the person being watched. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2017-111506 A Summary of the Invention [Problem to be solved by the invention]

[0004] Consider authenticating a moving object in a specified area. When multiple cameras are present in the specified area, conventionally, authentication of the moving object is performed for each of the camera images captured by the multiple cameras. When authentication is performed for camera images that do not include the moving object to be authenticated, the processing load increases.

[0005] An object of the present disclosure is to provide a technology capable of reducing the processing load when authenticating a mobile object in a predetermined area. [Means for solving the problem]

[0006] One aspect of the present disclosure relates to a mobile authentication system for authenticating a mobile object in a predetermined area. The mobile authentication system includes one or more processors capable of communicating with a plurality of cameras present in a predetermined area, and one or more storage devices for storing an authentication image of the mobile object and camera information indicating at least the position and orientation of each of the plurality of cameras. The ease of authenticating the mobile object depends at least on the distance between the mobile object and the camera, and increases as the distance decreases. When a request to authenticate the mobile object is received, the one or more processors acquire location information indicating the current location of the mobile object. Furthermore, the one or more processors select a target camera from the plurality of cameras whose ease of authentication is equal to or higher than a predetermined level based on at least the location information and the camera information. Furthermore, the one or more processors authenticate the mobile object based on a comparison between an image captured by the target camera and the authentication image. Effect of the Invention

[0007] According to the present disclosure, when a request to authenticate a moving object is received, a camera with a mobile object authentication ease of at least a predetermined level is selected as a target camera. Then, the moving object is authenticated based on a comparison between an image captured by the target camera and an authentication image. This allows the moving object to be authenticated with fewer cameras, thereby reducing the processing load. Furthermore, since an unnecessarily large number of images are not used for authentication, the opportunity for unnecessary authentication processing to be performed on people unrelated to the mobile object to be authenticated is reduced. This is preferable from the viewpoint of ensuring privacy. [Brief description of the drawings]

[0008] [Figure 1] FIG. 1 is a diagram for explaining an overview of a mobile authentication system according to a first embodiment. [Diagram 2] FIG. 4 is an explanatory diagram showing a specific example of the target camera selection process according to the first embodiment. [Diagram 3] FIG. 4 is an explanatory diagram showing a specific example of the recognizability improvement process according to the first embodiment; [Figure 4] 5 is a flowchart showing an example of a mobile object authentication process according to the first embodiment. [Diagram 5] FIG. 11 is an explanatory diagram showing a specific example of mobile object authentication processing according to the second embodiment. [Figure 6] FIG. 11 is an explanatory diagram showing a specific example of a mobile object authentication process according to the third embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] A mobile authentication system according to an embodiment of the present disclosure will be described with reference to the accompanying drawings. In addition, common elements in each drawing are given the same reference numerals and duplicated explanations will be omitted.

[0010] 1. First embodiment 1-1. Overview of the mobile authentication system FIG. 1 is a diagram for explaining an overview of a mobile object authentication system 1 according to a first embodiment. The mobile object authentication system 1 authenticates a mobile object 4 in a predetermined area 2. Examples of the mobile object 4 (mobile object 4A, mobile object 4B, etc.) to be authenticated include a human being, a vehicle, an animal, etc. For example, when the mobile object authentication system 1 is applied to a monitoring service, the mobile object 4 to be authenticated is a human being (e.g., a child) to be monitored. Examples of the predetermined area 2 include a region, a town, a building, etc. A request to authenticate the mobile object 4 may be issued from the mobile object 4 itself, or may be issued from a person (e.g., a parent) having a relationship with the mobile object 4 (e.g., a child). When an authentication request is issued from the mobile object 4 itself, for example, an operation is performed from a mobile terminal 5 (mobile terminal 5A, mobile terminal 5B, etc.) carried by the mobile object 4.

[0011] The mobile authentication system 1 includes a plurality of cameras 3 (camera 3A, camera 3B, camera 3C, etc.) present in a predetermined area 2, and a management server 10.

[0012] The management server 10 includes one or more processors 20 (hereinafter simply referred to as processor 20), one or more storage devices 30 (hereinafter simply referred to as storage devices 30), and a communication device 40. The processor 20 executes various processes. An example of the processor 20 is a CPU (Central Processing Unit). The storage device 30 stores various information required for processing by the processor 20. Examples of the storage device 30 include a volatile memory, a non-volatile memory, an HDD (Hard Disk Drive), and an SSD (Solid State Drive). The communication device 40 is capable of communicating with at least a plurality of cameras 3 present in a predetermined area 2 and a mobile terminal 5 carried by a moving object 4.

[0013] The mobile object authentication program (not shown) is a computer program executed by the processor 20. The processor 20 may execute the mobile object authentication program to realize various functions of the management server 10. The mobile object authentication program is stored in the storage device 30. Alternatively, the mobile object authentication program may be recorded in a computer-readable storage medium.

[0014] The various information stored in the storage device 30 includes camera information 31, position information 32, and an authentication image 33.

[0015] The camera information 31 includes information on the position and orientation of each of the multiple cameras 3 present in the specified area 2. When the position and orientation of a certain camera 3 are fixed, the information is given in advance as known information. When at least one of the position and orientation of a certain camera 3 is variable, the information is obtained by communication with that camera 3 via the communication device 40. The camera information 31 also includes a video database DB that accumulates video (images) captured by each of the multiple cameras 3. The video (images) captured by each camera 3 are obtained by communication with that camera 3 via the communication device 40.

[0016] The location information 32 indicates the current location of the mobile object 4 to be authenticated. The current location of the mobile object 4 is acquired, for example, by a GPS function of the mobile terminal 5 carried by the mobile object 4. As another example, the current location of the mobile object 4 may be estimated based on radio waves of a wireless LAN. The location information 32 of the mobile object 4 is obtained by communication with the camera 3 via the communication device 40. The location information 32 of the mobile object 4 may indicate a prediction result of the current location of the mobile object 4.

[0017] The authentication image 33 is an image of the moving body 4 that is registered in advance to authenticate the moving body 4. For example, when the moving body 4 to be authenticated is a person, the authentication image 33 may be the face of that person. As another example, the authentication image 33 may be an entire image of the moving body 4 to be authenticated. Such an authentication image 33 is registered in advance by the moving body 4 itself or a person related to the moving body 4.

[0018] The outline of the mobile object authentication process by the processor 20 (management server 10) according to the present embodiment is as follows. When a request to authenticate a mobile object 4 is received, the processor 20 selects a "target camera" to be used to authenticate the mobile object 4 from among the multiple cameras 3. At this time, the processor 20 selects the target camera based on the "ease of authentication" of the mobile object 4. The ease of authentication of the mobile object 4 depends at least on the distance between the mobile object 4 and the camera 3. The shorter the distance between the mobile object 4 and the camera 3, the higher the ease of authentication of the mobile object 4. Details of the ease of authentication of the mobile object 4 will be described later. The processor 20 acquires the above-mentioned position information 32 and camera information 31. Furthermore, the processor 20 selects a target camera from among the multiple cameras 3 whose ease of authentication is equal to or higher than a predetermined level based on at least the position information 32 and the camera information 31. Then, the processor 20 authenticates the mobile object 4 based on a comparison between an image captured by the target camera and an authentication image 33 of the mobile object 4.

[0019] Management information 34 is information used to manage mobile object authentication processing. For example, management information 34 indicates the correspondence between a mobile object 4 and a target camera used to authenticate the mobile object 4. Management information 34 is also stored in storage device 30. In the example shown in Fig. 1, camera 3A is selected as the target camera used to authenticate mobile object 4A, and camera 3C is selected as the target camera used to authenticate mobile object 4B.

[0020] As described above, according to this embodiment, when a request to authenticate a moving object 4 is received, a camera 3 for which the ease of authentication of the moving object 4 is at or above a predetermined level is selected as a "target camera". Then, the moving object 4 is authenticated based on a comparison between an image captured by the target camera and the authentication image 33. This allows the moving object 4 to be authenticated with fewer cameras, thereby reducing the processing load. Furthermore, because an unnecessarily large number of images are not used for authentication, the chances of performing unnecessary authentication processing on people unrelated to the mobile object 4 to be authenticated are reduced. This is preferable from the viewpoint of ensuring privacy.

[0021] 1-2. Example of target camera selection The management server 10 selects a target camera from among the multiple cameras 3 based on the ease of authenticating the moving object 4. The ease of authenticating the moving object 4 depends at least on the distance between the moving object 4 and the camera 3. The shorter the distance between the moving object 4 and the camera 3, the easier it is to authenticate the moving object 4. Therefore, the management server 10 may select one of the multiple cameras 3 that is closest to the moving object 4 as the target camera. More specifically, the management server 10 calculates the distance between the moving object 4 and each camera 3 based on the camera information 31 and the position information 32 of the moving object 4. Then, the management server 10 may select the camera 3 with the shortest distance as the target camera.

[0022] In addition to the distance between the mobile object 4 to be authenticated and the camera 3, other parameters may be taken into account.

[0023] FIG. 2 is a diagram for explaining a specific example of the selection of a target camera. In the example shown in FIG. 2, the ease of authentication of a moving object 4 is represented by an "evaluation score." The evaluation score is calculated based on various parameters that affect the ease of authentication of the moving object 4. The various parameters that affect the ease of authentication of the moving object 4 include, for example, a first parameter P1, a second parameter P2, and a third parameter P3. The first parameter P1 is the distance between the moving object 4 to be authenticated and the camera 3. The second parameter P2 is the relationship between the position of the moving object 4 and the orientation of the camera 3, and indicates whether the moving object 4 to be authenticated is present within the angle of view of the camera 3. The third parameter P3 indicates the shooting situation when the camera 3 shot the moving object 4 to be authenticated.

[0024] A calculation example of the evaluation value of each parameter will be described. In calculating the first parameter P1, the management server 10 calculates the distance between the moving object 4 and the camera 3 based on the information on the positions of the multiple cameras 3 included in the camera information 31 and the position information 32 of the moving object 4. Then, the shorter the distance, the higher the management server 10 sets the first parameter P1. For example, the management server 10 determines whether the distance is less than a predetermined distance. If the distance is less than the predetermined distance, that is, if the distance is short, the management server 10 sets the evaluation value of the first parameter P1 to be high. On the other hand, if the distance is equal to or greater than the predetermined distance, that is, if the distance is long, the management server 10 sets the evaluation value of the first parameter P1 to be low. The threshold may be set in multiple stages. Also, the first parameter P1 may change continuously according to the distance.

[0025] In calculating the second parameter P2, the management server 10 determines whether or not the moving object 4 is present within the angle of view of each camera 3 based on the information on the orientation of each of the multiple cameras 3 included in the camera information 31 and the position information 32 of the moving object 4. Then, the management server 10 sets a high value of the second parameter P2 corresponding to the camera 3 for which it is determined that the moving object 4 is present within the angle of view. On the other hand, the management server 10 sets a low value of the second parameter P2 corresponding to the camera 3 for which it is determined that the moving object 4 is present outside the angle of view. In other words, the value of the second parameter P2 is high when the moving object 4 to be authenticated is present within the angle of view, and is low when the moving object 4 is present outside the angle of view. The management server 10 may set the value of the second parameter P2 according to the position of the moving object 4 present within the angle of view of the camera 3. For example, the management server 10 may set the value of the second parameter P2 higher as the position of the moving object 4 within the angle of view of the camera 3 is closer to the center of the horizontal angle of view. Note that a camera 3 in which a moving object 4 is outside the angle of view is not selected as a target camera, and therefore the evaluation value may be set to, for example, a negative value. Also, as described later, if the position or orientation of the camera 3 is variable, the position or orientation of the camera 3 may be changed so that the moving object 4 is within the angle of view.

[0026] In calculating the third parameter P3, the management server 10 recognizes the shooting conditions for each camera 3 based on the images for each camera 3 included in the video database DB of the camera information 31. The shooting conditions include at least one of illuminance, backlight, weather, and object density in the image. Then, the management server 10 judges whether the shooting conditions recognized for each camera 3 satisfy the shooting possible conditions. If it is judged that the shooting conditions satisfy the shooting possible conditions, the management server 10 sets a high evaluation value of the third parameter P3 corresponding to the camera 3 that satisfies the shooting possible conditions. On the other hand, if it is judged that the shooting conditions do not satisfy the shooting possible conditions, the management server 10 sets a low evaluation value of the third parameter P3 corresponding to the camera 3 that does not satisfy the shooting possible conditions. In other words, the better the shooting conditions, the higher the value of the third parameter P3, and the worse the shooting conditions, the lower the value.

[0027] The shooting conditions are set by taking into consideration at least one of the following, for example: the brightness of the image is within a specified range; the degree of whiteout in the image is small; the image has sufficient visibility to detect buildings, people, etc.; and the object density in the image is below a threshold value.

[0028] An example of calculating an evaluation score based on the evaluation value for each parameter will be described below. In a first calculation method, the evaluation value is the sum of the evaluation values ​​for each parameter (see FIG. 2(A)).

[0029] In the second calculation method, the importance of each parameter is preset and taken into consideration (see FIG. 2(B)). The importance is expressed as, for example, "high", "medium", or "low". In the example shown in FIG. 2(B), the importance of the first parameter P1 and the second parameter P2 is "high", and the importance of the third parameter P3 is "medium". Then, the product of the evaluation value and the importance is calculated for each parameter, and the sum of the products for each parameter is calculated as the evaluation score. Alternatively, a weighted average of the evaluation values ​​for each parameter may be calculated as the evaluation score. According to the second calculation method, the importance of each parameter is taken into consideration, so that the evaluation score can be calculated appropriately.

[0030] The higher the evaluation score, the easier it is to authenticate the moving object 4, and the lower the evaluation score, the easier it is to authenticate the moving object 4 (see FIG. 2(C)). After calculating the evaluation score, the management server 10 selects a target camera whose evaluation score (i.e., the ease of authenticating the moving object 4) is equal to or higher than a predetermined level. The predetermined level may be a predetermined threshold value, or may be a different threshold value for each type of moving object 4 (human, vehicle, animal, etc.). For example, if the predetermined level is "60", in the example shown in FIG. 2(A), the management server 10 selects the camera 3A and the camera 3B whose evaluation score is equal to or higher than the predetermined level as the target camera. On the other hand, in the example shown in FIG. 2(B), the management server 10 selects the camera 3A whose evaluation score is equal to or higher than the predetermined level as the target camera. Regardless of the evaluation score, the camera 3 in which the moving object 4 is outside the angle of view is not selected as the target camera.

[0031] In addition, when there are multiple cameras 3 with evaluation scores equal to or higher than a predetermined level, one camera 3 may be further selected from the multiple cameras 3. Specifically, the management server 10 may select one of the multiple cameras 3 that is most likely to authenticate a moving object 4 as the target camera.

[0032] In this manner, at least one camera 3 is selected from the multiple cameras 3 as the target camera used for authenticating the moving body 4. The ease of authenticating the moving body 4 (evaluation score) is used to select the target camera. By selecting a camera 3 with a high ease of authenticating the moving body 4 as the target camera, the moving body 4 can be authenticated with fewer cameras, thereby reducing the processing load.

[0033] 1-3. Examples of improvements to ease of authentication If there is no camera 3 for which the ease of authentication of the moving body 4 is equal to or higher than a predetermined level, a target camera to be used for authenticating the moving body 4 will not be selected. Therefore, the management server 10 may implement a process for actively improving the ease of authentication of the moving body 4. Fig. 3 is a diagram for explaining a specific example of a process for improving the ease of authentication of the moving body 4. Three examples of a process for improving the ease of authentication of the moving body 4 will be given.

[0034] The first example is a case where the moving object 4 does not exist within the angle of view of the camera 3 closest to the moving object 4, as shown in FIG. 3(A). In this case, the ease of authenticating the moving object 4 decreases. However, the camera 3 may be a "variable orientation camera" whose orientation is variable. In that case, for example, as shown in FIG. 3(B), the orientation of the camera 3 (variable orientation camera) may be changed so that the moving object 4 exists within the angle of view of the camera 3. Note that even if the orientation of the camera 3 closest to the moving object 4 is fixed, if the multiple cameras 3 include a variable orientation camera, the orientation of the variable orientation camera may be changed so that the moving object 4 exists within the angle of view of the variable orientation camera.

[0035] According to the first example, the management server 10 calculates an angle for rotating the variable-direction camera so as to align the direction of the variable-direction camera with the position of the moving object 4, based on the direction information of the variable-direction camera included in the camera information 31 and the position information 32 of the moving object 4. Then, the management server 10 transmits rotation instruction information to the variable-direction camera via the communication device 40, which instructs the direction of the variable-direction camera to rotate based on the calculated angle. This brings the moving object 4 into the angle of view of the variable-direction camera, improving the ease of authentication of the moving object 4. If the ease of authentication of the moving object 4 by the variable-direction camera is equal to or higher than a predetermined level, the variable-direction camera is selected as a target camera.

[0036] The second example is a case where the distance from the fixed-position camera 3 to the moving object 4 is long, as shown in FIG. 3(C). In this case, the moving object 4 included in the image captured by the camera 3 becomes unclear, and the ease of authenticating the moving object 4 decreases. However, there may be a case where a "position-variable camera" whose position is variable can be used. Examples of position-variable cameras include an in-vehicle camera and a camera mounted on a drone. When a position-variable camera is available, the position of the position-variable camera may be changed so that the moving object 4 is within the angle of view of the position-variable camera, as shown in FIG. 3(D). When there are multiple position-variable cameras, the position of the position-variable camera closest to the moving object 4 may be changed.

[0037] According to the second example, the management server 10 transmits movement instruction information to the position variable camera via the communication device 40, based on the position information of the position variable camera included in the camera information 31 and the position information 32 of the moving body 4, so as to move the position of the position variable camera closer to the position of the moving body 4. This makes it possible to photograph the moving body 4 from a closer position, improving the ease of authentication of the moving body 4. If the ease of authentication of the moving body 4 by the variable camera is equal to or higher than a predetermined level, the position variable camera is selected as the target camera.

[0038] The second example is also applicable to the situation shown in Fig. 3(A). In that case, the management server 10 transmits a movement instruction to the position variable camera so that the moving object 4 is included in the angle of view of the position variable camera. This brings the moving object 4 into the angle of view of the position variable camera, improving the ease of authentication of the moving object 4. If the ease of authentication of the moving object 4 by the position variable camera reaches or exceeds a predetermined level, the position variable camera is selected as the target camera.

[0039] As explained in the first and second examples, by changing at least one of the camera position and camera direction of the movable camera, it is possible to improve the ease of authenticating the moving object 4.

[0040] The third example is a case where the moving object 4 is present within the angle of view of the camera 3 closest to the moving object 4, and the image IMG captured by the camera 3 has low image quality, as shown in FIG. 3(E). For example, the image IMG captured in a dark or backlit environment has low image quality. In this case, the moving object 4 included in the image captured by the camera 3 becomes unclear, and the ease of authenticating the moving object 4 decreases. In this case, the management server 10 may improve the image quality of the image IMG so that the ease of authenticating the moving object 4 increases. For example, as shown in FIG. 3(F), the image quality of the image IMG can be improved by performing a brightness correction process. As a result, the moving object 4 to be authenticated included in the image IMG becomes clear, and the ease of authenticating the moving object 4 improves.

[0041] 1-4. Processing flow example FIG. 4 is a flowchart briefly illustrating an example of a mobile authentication process according to the first embodiment.

[0042] In step S100, the management server 10 determines whether or not a request to authenticate the mobile object 4 has been received. If a request to authenticate the mobile object 4 has been received (step S100; Yes), the process proceeds to step S110. Otherwise (step S100; No), the process ends.

[0043] In step S110, the management server 10 acquires the location information 32 of the moving object 4. After that, the process proceeds to step S120.

[0044] In step S120, the management server 10 calculates the ease of authentication of the moving object 4 for each camera 3. After that, the process proceeds to step S130.

[0045] In step S130, the management server 10 determines whether the ease of authentication of the mobile object 4 is equal to or higher than a predetermined level. If the ease of authentication of the mobile object 4 is equal to or higher than the predetermined level (step S130; Yes), the process proceeds to step S140. Otherwise (step S130; No), the process proceeds to step S160.

[0046] In step S140, the management server 10 selects, as a target camera, a camera 3 for which the ease of authenticating the moving object 4 is equal to or higher than a predetermined level. After that, the process proceeds to step S150.

[0047] In step S150, the management server 10 authenticates the moving object 4 based on a comparison between the image captured by the target camera and the authentication image 33.

[0048] In step S160, the management server 10 improves the ease of authentication of the mobile entity 4 so that the ease of authentication of the mobile entity 4 is equal to or higher than a predetermined level. Then, the process returns to step S130.

[0049] Effects According to the first embodiment, when a request to authenticate a moving object 4 is received, a camera 3 for which the ease of authentication of the moving object 4 is at or above a predetermined level is selected as a "target camera". Then, the moving object 4 is authenticated based on a comparison between an image captured by the target camera and the authentication image 33. This allows the moving object 4 to be authenticated with fewer cameras, thereby reducing the processing load. Furthermore, because an unnecessarily large number of images are not used for authentication, the chances of unnecessary authentication processing being performed on people unrelated to the mobile object 4 to be authenticated are reduced. This is preferable from the viewpoint of ensuring privacy.

[0050] 2. Second embodiment In the second embodiment, a method for improving the ease of authentication of a mobile body 4 by actively influencing the mobile body 4 to be authenticated will be described. In the second embodiment, the mobile body 4 to be authenticated is a human who can comply with a request from the management server 10. The management server 10 requests the human to be authenticated to perform a predetermined action. Specifically, the management server 10 transmits instruction information requesting the human to perform a predetermined action via the communication device 40 to the portable terminal 5 carried by the human.

[0051] In the example shown in FIG. 5(A), the face of the person to be authenticated is not visible from the camera 3, and the authentication process fails. In this case, the management server 10 may request the person to be authenticated to change his / her orientation. That is, the predetermined action may be to change his / her orientation. As another example, when the camera 3 is installed at a nearby landmark, the management server 10 may request the person to be authenticated to face the landmark. That is, the predetermined action may be to face the landmark. As shown in FIG. 5(B), there is a possibility that the person's face becomes visible from the camera 3 by changing his / her orientation. As a result, there is a possibility that the authentication process is successful.

[0052] In the example shown in FIG. 5(C), many moving objects 4 are captured in the image captured by the camera 3. In this case, the management server 10 may request the person to be authenticated to perform a predetermined motion so as to make it easier to distinguish the person from other moving objects 4. Examples of the predetermined motion include waving hands and jumping. As shown in FIG. 5(D), the person who receives the request performs the predetermined motion. The management server 10 performs image analysis to detect the object performing the predetermined motion. The management server 10 then performs authentication processing on the detected object.

[0053] In this way, by requesting the person to be authenticated to perform a predetermined action, it becomes easier to identify the person to be authenticated in the image IMG. As a result, the ease of authenticating the moving object 4 is improved. Also, the accuracy of authenticating the moving object 4 is improved.

[0054] A combination of the first and second embodiments is also possible.

[0055] 3. Third embodiment For example, consider a case where the density of people who are not the authentication target is high around the mobile object 4 to be authenticated. In this case, the image captured by the target camera includes people who are not related to the mobile object 4 to be authenticated. In this case, unnecessary authentication processing is performed on people who are not related to the mobile object 4. This is not preferable from the viewpoint of ensuring privacy. Therefore, in the third embodiment, the area in which the mobile object 4 is expected to appear in the image captured by the target camera is narrowed down.

[0056] FIG. 6 is an explanatory diagram showing a specific example of the mobile object authentication process according to the third embodiment. The management server 10 acquires an image IMG captured by a target camera. The position and orientation of the target camera are included in the camera information 31. The position of the mobile object 4 is obtained from the position information 32. Therefore, the management server 10 can narrow down the area in the image IMG where the mobile object 4 to be authenticated is assumed to be captured, based on the camera information 31 and the position information 32 of the mobile object 4. The image IMG of the narrowed down area is called a partial area image IMG_T. The range of the partial area image IMG_T is smaller than the range of the original image IMG. In the example shown in FIG. 6(A), the management server 10 trims the partial area image IMG_T from the original image IMG. Then, the management server 10 authenticates the mobile object 4 based on a comparison between the partial area image IMG_T and the authentication image 33 of the mobile object 4.

[0057] In the example shown in Fig. 6(B), the management server 10 instructs the target camera to zoom in on an area in which the moving object 4 to be authenticated is assumed to appear. As described above, the area in the original image IMG in which the moving object 4 is assumed to appear can be estimated based on the camera information 31 and the position information 32 of the moving object 4. The management server 10 acquires the zoomed-in image from the target camera as a partial area image IMG_T. Then, the management server 10 authenticates the moving object 4 based on a comparison between the partial area image IMG_T and an authentication image 33 of the moving object 4.

[0058] According to the third embodiment, the partial area image IMG_T in which the moving object 4 to be authenticated is assumed to appear is narrowed down from the image IMG captured by the target camera. Then, the moving object 4 is authenticated based on a comparison between the partial area image IMG_T and the authentication image 33. Compared to the original image IMG, the probability that a person unrelated to the moving object 4 to be authenticated appears in the partial area image IMG_T is greatly reduced. Therefore, the chances of performing unnecessary authentication processing on a person unrelated to the moving object 4 to be authenticated are reduced. This is preferable from the viewpoint of ensuring privacy. Furthermore, by using the partial area image IMG_T, it becomes easier to identify the moving object 4, thereby improving the accuracy of authentication of the moving object 4.

[0059] A combination of the first embodiment and the third embodiment, or a combination of the second embodiment and the third embodiment, is also possible. [Explanation of symbols]

[0060] 1...mobile object authentication system, 2...predetermined area, 3, 3A, 3B, 3C...camera, 4, 4A, 4B...mobile object, 5, 5A, 5B...mobile terminal, 10...management server, 20...processor, 30...storage device, 31...camera information, 32...location information, 33...image for authentication, 34...management information, 40...communication device, IMG...image, IMG_T...partial region image

Claims

1. A mobile object authentication system for authenticating a mobile object in a predetermined area, comprising: One or more processors capable of communicating with a plurality of cameras present in the predetermined area; one or more storage devices that store an authentication image of the moving object and camera information indicating at least a position and an orientation of each of the plurality of cameras; Including, The ease of authenticating the moving object depends at least on the distance between the moving object and the camera, and increases as the distance decreases; When a request is received to authenticate the mobile entity, the one or more processors: acquiring location information indicating a current location of the moving object; selecting a target camera having the ease of authentication equal to or higher than a predetermined level from among the plurality of cameras based on at least the position information and the camera information; The moving object is authenticated based on a comparison between the image captured by the target camera and the authentication image. It was configured as Mobile authentication system.

2. 2. The mobile authentication system according to claim 1, The one or more processors select, based on the position information and the camera information, a camera among the plurality of cameras that is closest to the moving object as the target camera, or select, from the plurality of cameras, a camera that is the easiest to authenticate as the target camera. Mobile authentication system.

3. 2. The mobile authentication system according to claim 1, the plurality of cameras includes a variable camera in which at least one of a camera position and a camera orientation is variable, The one or more processors: changing at least one of the camera position and the camera orientation of the variable camera so as to increase the ease of authentication regarding the variable camera; Taking into consideration a change in at least one of the camera position and the camera direction of the movable camera, the target camera having the ease of authentication equal to or higher than the predetermined level is selected. Mobile authentication system.

4. 2. The mobile authentication system according to claim 1, The moving entity is a human being, The one or more processors further include: Requesting the human to perform a predetermined action. Mobile authentication system.

5. 2. The mobile authentication system according to claim 1, The one or more processors: Narrowing down partial area images in which the moving object is assumed to appear in the image based on the position information and the camera information; The moving object is authenticated based on a comparison between the partial area image and the authentication image. Mobile authentication system.

Images