Control device and control method

Abstract

To speed up face matching processing used to manage the entry and exit of persons passing through gates.SOLUTION: A control device includes: an acquisition unit for acquiring an image containing a face image area of a person entering a gate whose passage eligibility is determined based on the result of face matching processing, taken from a direction from an exit to an entrance of the gate; and a control unit that determines whether or not to perform the face matching processing on the person based on the illuminance and / or blur of the person detected in the image.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present disclosure relates to a control device and a control method. [Background technology]

[0002] Technology is being considered that uses face matching processing to manage the entry and exit of people passing through gates installed at stations, airports, etc. Patent Document 1 describes a method of detecting people from images captured around the gate and determining the control details for each gate based on the feature amounts in the images of the people. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2005-070850 A Summary of the Invention [Problem to be solved by the invention]

[0004] There is room for improvement in speeding up the face matching process used to manage the entry and exit of people passing through gates.

[0005] A non-limiting embodiment of the present disclosure contributes to providing a control device and a control method that realizes high-speed face matching processing used in managing the entry and exit of people passing through a gate. [Means for solving the problem]

[0006] A control device according to one embodiment of the present disclosure includes an acquisition unit that acquires an image including a facial image area of ​​a person entering a gate, where passage or non-passage is determined based on the results of facial matching processing, photographed from a direction from the exit of the gate toward the entrance, and a control unit that determines whether or not to perform the facial matching processing of the person based on the illuminance and / or blur of the person detected from the image.

[0007] In a control method according to one embodiment of the present disclosure, a control device acquires an image including a facial image area of ​​a person entering a gate, where passage or non-passage is determined based on the results of facial matching processing, photographed from a direction from the exit of the gate toward the entrance, and determines whether or not to perform the facial matching processing of the person based on the illuminance and / or blur of the person detected in the image.

[0008] In addition, these comprehensive or specific aspects may be realized by a system, an apparatus, a method, an integrated circuit, a computer program, or a recording medium, or may be realized by any combination of a system, an apparatus, a method, an integrated circuit, a computer program, and a recording medium. Effect of the Invention

[0009] A non-limiting embodiment of the present disclosure can realize high-speed face matching processing used in managing the entry and exit of people passing through a gate.

[0010] Further advantages and benefits of an embodiment of the present disclosure will become apparent from the specification and drawings, in which such advantages and / or benefits are provided by some of the embodiments and features described in the specification and drawings, respectively, but not necessarily all of which are provided to obtain one or more identical features. [Brief description of the drawings]

[0011] [Figure 1] A diagram showing an example of ticket gate control using face recognition [Diagram 2] FIG. 1 is a diagram showing an example of a system configuration of a currency management system according to an embodiment; [Diagram 3] A diagram showing an example of the movement of a person to be verified who intends to pass through [Figure 4] A diagram showing an example of the movement of a person to be verified who has no intention of passing through [Diagram 5] A flowchart showing an example of the flow of face matching processing in one embodiment. [Figure 6] A flowchart showing the flow of the passage intention determination process in S105 of FIG. [Figure 7] FIG. 13 is a diagram showing an example of parameters used in determining whether a user intends to pass through the vehicle; [Figure 8] FIG. 13 is a diagram showing an example of an interrupt determination; [Figure 9] FIG. 13 is a diagram showing an example of the relationship between the face matching start position and brightness. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] Hereinafter, preferred embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. In this specification and the drawings, components having substantially the same functions are denoted by the same reference numerals, and redundant description will be omitted.

[0013] (One embodiment) <Knowledge that led to this disclosure> Fig. 1 is a diagram showing an example of ticket gate control using face recognition. Fig. 1 shows control at a ticket gate installed at a station and control performed by a server. The control at the ticket gate is executed, for example, by a control device connected to the ticket gate. The server also stores information (for example, facial feature values) of people registered in advance (hereinafter, "registered persons") and performs face matching in response to a request from the ticket gate.

[0014] Note that face matching corresponds to a process of comparing the face of a person in an image with the faces of registered persons to identify which of the registered persons the person in the image is. Face authentication corresponds to a process of performing face matching and determining whether or not the registered person identified by face matching is allowed to pass through. However, since face matching is performed in face authentication, in the following, "face authentication" and "face matching" may be interchangeable.

[0015] A camera is provided in the ticket gate. The camera captures an image including the face of a person present around the ticket gate. The person present around the ticket gate includes a person approaching the ticket gate to pass through the ticket gate, a person passing around the ticket gate without passing through the ticket gate, a person remaining around the ticket gate, and a person walking around the ticket gate.

[0016] The camera captures an image (S11). The camera may capture images all the time, or may capture images only when a person is detected near the ticket gate.

[0017] The control device of the ticket gate detects a face frame in the image captured in S11 (S12). For example, if the captured image includes a human face, the face frame corresponds to a rectangular frame that surrounds the face.

[0018] The control device of the ticket gate performs face matching processing (S13). For example, the control device extracts information to be used in face matching processing from the face image included in the detected face frame, and requests face matching from the server by sending the extracted information to the server. The control device then obtains a response to the face matching request from the server. Here, the information extracted from the face image included in the detected face frame includes facial features. Also, the person in the face image included in the detected face frame will be referred to as the "matching target person" below.

[0019] The server performs face matching in response to the face matching request from the ticket gate in S13 (S21). For example, the server matches information used for face matching processing obtained from the ticket gate (e.g., facial feature amount) with information of registered persons stored in the server (e.g., facial feature amount), and identifies which registered persons are the persons to be matched.

[0020] In S21, when it is determined that the person to be verified is one of the registered persons, it is determined whether the person to be verified corresponding to the identified registered person has the right to pass (S23). For example, when the identified registered person has purchased a ticket, it is determined that the person to be verified corresponding to the identified registered person has the right to pass.

[0021] In addition, if it is not possible to identify which of the registered persons the person to be matched is, or if it is possible to identify which of the registered persons the person to be matched is, but the identified registered person does not have the right to pass, it is determined that the person to be matched does not have the right to pass.

[0022] The server transmits a response including the result of the face matching in S21 to the control device of the ticket gate. For example, the response transmitted from the server to the control device of the ticket gate includes a determination result indicating whether the person being matched has the right to pass.

[0023] The control device of the ticket gate controls passage through the ticket gate after acquiring a response to the face matching request from the server (S14). For example, when the acquired response includes a determination result that the person being matched has the right to pass, the control device controls to permit passage through the ticket gate. The control to permit passage through the ticket gate includes control to open a door provided in the passage of the ticket gate, and control to output audio and / or display to notify the person being matched that he / she has been permitted to pass. In addition, when the acquired response includes a determination result that the person being matched does not have the right to pass, the control device controls to block passage through the ticket gate. The control to block passage through the ticket gate includes control to close a door provided in the passage of the ticket gate, and control to output audio and / or display to notify the person being matched that he / she has been prohibited from passing.

[0024] In the ticket gate control using face recognition illustrated in FIG. 1, when the image captured by the camera includes a person's face image, face matching processing is performed on the face image. In this control, face matching processing is also performed on face images of people who do not actually pass through the ticket gate. A person who does not actually pass through the ticket gate corresponds to a person who has no intention of passing through the ticket gate. Specifically, this includes a person who passes in front of the ticket gate without entering the ticket gate, a person who stops in front of the ticket gate, a person who once entered the ticket gate but then made a U-turn and left the ticket gate, and the like. The face matching processing on the face images of people who have no intention of passing through the ticket gate increases the processing time of the face matching processing in the ticket gate control.

[0025] For example, by performing face matching processing on a face image of a person who has no intention of passing through a ticket gate, control that allows the person to pass through the ticket gate or control that blocks the person from passing through the ticket gate may be performed unnecessarily, even though the person will not actually pass through the ticket gate. For example, if a ticket gate allows passage in both directions, this unnecessary control may prevent passage from the side on which the person who has no intention of passing through the ticket gate is present and the opposite side. Furthermore, this unnecessary control may delay processing (e.g., face matching processing) for a person who has the intention of passing through the ticket gate after a person who has no intention of passing through the ticket gate.

[0026] In addition, when control to allow or block the passage of a person who does not actually pass through the ticket gate is performed unnecessarily, the person may be erroneously charged for passing through the ticket gate. In addition, as shown in Fig. 1, when communication is performed between the control device of the ticket gate and the server, face matching processing is performed on the face image of a person who has no intention of passing through the ticket gate, which increases communication traffic.

[0027] Therefore, in this embodiment, it is determined whether or not a person included in an image captured by a camera has an intention to pass through the ticket gate, and face matching processing is omitted for face images of people who are determined to have no intention to pass through the ticket gate. Here, omission of face matching processing means control in general such that the ticket gate does not perform face matching processing, and includes skipping to the next process without executing the face matching processing function itself, and / or stopping the face matching processing that is proceeding in parallel with the determination of whether or not there is an intention to pass through. This makes it possible to concentrate computer resources on the face matching processing of people who are determined to have an intention to pass through the ticket gate, reducing the processing time of the face matching processing in the ticket gate control, and realizing high-speed face matching processing.

[0028] In addition, a person's intention to pass through a ticket gate may correspond to the person having plans to pass through the ticket gate, the person intending to pass through the ticket gate within a specified time, or the probability that the person will pass through the ticket gate is high (e.g., above a threshold value).

[0029] In addition, a determination that a person has an intention to pass through a ticket gate corresponds to a determination to execute face matching processing for that person, and a determination that a person has no intention to pass through a ticket gate corresponds to a determination not to execute face matching processing for that person. For example, a determination of whether or not there is an intention to pass through corresponds to a determination of whether or not there is a face matching processing to be executed.

[0030] <An example of system configuration> Fig. 2 is a diagram showing an example of the system configuration of the passage management system 1 according to the present embodiment. Fig. 2 shows a camera 10, a control device 20, a server device 30, and a ticket gate 40. The ticket gate 40 is connected to the control device 20 by wire and / or wirelessly. The camera 10 is connected to the control device 20 by wire and / or wirelessly. The server device 30 may be connected to the control device 20 by wire and / or wirelessly, or may be connected via a network (e.g., the Internet).

[0031] A camera 10 is provided in the ticket gate 40. The camera 10 captures an image including, for example, the face of a person present around the ticket gate 40 from a direction from the exit to the entrance of the ticket gate 40 (arrow R in FIG. 2). For example, people present around the ticket gate 40 include people who are trying to pass through the ticket gate 40 and coming towards the ticket gate 40, people who pass around the periphery of the ticket gate 40 without passing through the ticket gate 40, people who remain around the ticket gate 40, and people who come and go around the periphery of the ticket gate 40.

[0032] A control device 20 is connected to the ticket gate 40. The control device 20 controls the ticket gate 40 in accordance with the result of face matching processing based on the image captured by the camera 10.

[0033] The control device 20 includes a communication unit 201, an intention determination processing unit 202, a face matching processing unit 203, and a ticket gate control unit 204. Note that the intention determination processing unit 202, the face matching processing unit 203, and the ticket gate control unit 204 may be collectively referred to as a control unit (or a processing unit).

[0034] The communication unit 201 performs communication between the control device 20 and the camera 10, and between the control device 20 and the server device 30. For example, the communication unit 201 acquires a captured image from the camera 10. The communication unit 201 also transmits information indicating a request for face matching to the server device 30, and receives a response to the request for face matching from the server device 30.

[0035] The intention determination processing unit 202 performs face detection processing on the captured image acquired from the camera 10. When a face frame included in the captured image is detected by the face detection processing, the intention determination processing unit 202 detects parameters from the face image within the face frame. The detected parameters are, for example, parameters related to the posture of the person to be matched relative to the ticket gate 40 and / or parameters related to the position of the person to be matched relative to the ticket gate 40.

[0036] The intention determination processing unit 202 performs a passing intention determination process based on the detected parameters, and determines whether or not the person to be matched corresponding to the face frame has the intention to pass through the ticket gate 40. In other words, the intention determination processing unit 202 determines whether or not to perform face matching processing of the person to be matched corresponding to the face frame based on parameters detected from the captured image. The passing intention determination process will be described later.

[0037] The face matching processing unit 203 performs face matching processing when a person to be matched who corresponds to the face frame intends to pass through the ticket gate 40. For example, the face matching processing unit 203 generates information indicating a request for face matching including face feature amounts extracted from the face image of the person to be matched, and transmits the information to the server device 30 via the communication unit 201.

[0038] The face matching processing unit 203 acquires a response to the request for face matching via the communication unit 201. The response to the request for face matching includes the determination result of the server device 30 as to whether or not the person to be matched has the right to pass through the ticket gate 40.

[0039] The ticket gate control unit 204 controls the ticket gate depending on whether the person being verified has the right to pass. When the acquired response includes a determination result that the person being verified has the right to pass, the ticket gate control unit 204 performs control to permit the person being verified to pass through the ticket gate 40. The control to permit the person being verified to pass through the ticket gate 40 includes control to open a door provided in the passage of the ticket gate 40, and control to output a voice and / or a display to notify the person being verified that he / she has been permitted to pass. Furthermore, when the acquired response includes a determination result that the person being verified does not have the right to pass, the ticket gate 40 performs control to block the person being verified to pass through the ticket gate 40. The control to block the person being verified to pass through the ticket gate 40 includes control to close a door provided in the passage of the ticket gate 40, and control to output a voice and / or a display to notify the person being verified that he / she has been prohibited from passing through.

[0040] The server device 30 includes a storage unit 301, a face matching unit 302, and a communication unit 303. The server device 30 may be a cloud server.

[0041] The storage unit 301 stores information about registrants, including face images acquired from the registrants, in advance. A registrant is a person who registers information in order to receive a service that allows the registrant to pass through the ticket gate 40 based on the result of face matching processing. The registered information includes facial features of the registrant. The storage unit 301 also stores information about whether or not each of the registrants has the right to pass.

[0042] The face matching unit 302 acquires a request for face matching from the control device 20 via the communication unit 203, and performs face matching. The face matching unit 302 matches the facial feature amount of the person to be matched included in the request for face matching with the facial feature amount of the registered person stored in the storage unit 301, and calculates a score indicating the similarity between the facial feature amount of the person to be matched and the facial feature amount of the registered person for each registered person. Then, if the highest score is equal to or greater than a threshold, the face matching unit 302 specifies that the person to be matched is the registered person corresponding to the highest score. When the face matching unit 302 specifies which of the registered people the person to be matched is, it determines whether the specified registered person has the right to pass. For example, if the specified registered person has purchased a ticket, it determines that the person to be matched corresponding to the specified registered person has the right to pass.

[0043] In addition, if the face matching unit 302 is unable to identify which of the registered persons the person to be matched is, or if it is able to identify which of the registered persons the person to be matched is but the identified registered person does not have the right to pass, it determines that the person to be matched does not have the right to pass.

[0044] The face matching unit 302 transmits a response including the result of face matching to the control device 20. For example, the response includes a determination result indicating whether or not the person being matched has the right of passage.

[0045] In the system configuration shown in FIG. 2, the control device 20 and the server device 30 are separate devices, but the present disclosure is not limited to this. The control device 20 may be configured integrally with the server device 30. In addition, in the system configuration shown in FIG. 2, the control device 20 is connected to the ticket gate 40 by wire or wirelessly, but the control device 20 may be built into the ticket gate 40. In addition, in the passage management system 1 according to the present embodiment, a part of the configuration shown in FIG. 2 may be omitted, or a configuration not shown in FIG. 2 may be added. For example, the ticket gate 40 may be equipped with lighting. The brightness, color, irradiation range, irradiation angle, etc. of the installed lighting may be fixed, or may be controlled by the ticket gate control unit 204 of the control device 20.

[0046] <Example of Passage Intention Determination Processing> An example of the passing intention determination process will be described. It is assumed that a person who has the intention to pass will gradually approach the ticket gate 40. On the other hand, it is assumed that a person who has no intention to pass will not approach the ticket gate 40 but will remain stationary or move around the vicinity. In this embodiment, therefore, the presence or absence of the person to be matched having the intention to pass is determined based on a parameter related to the position of the person to be matched relative to the ticket gate 40. In other words, in this embodiment, it is determined whether or not to perform face matching processing of the person to be matched based on a parameter related to the position of the person to be matched relative to the ticket gate 40. Then, in this embodiment, the control device 20, for example, detects illuminance and / or blur from the face image and determines the presence or absence of the intention to pass based on the change in illuminance and / or blur.

[0047] Fig. 3 is a diagram showing an example of the movement of a person to be verified who has the intention to pass through. Fig. 3 shows a time series of images taken by camera 10 installed in ticket gate 40 when the person to be verified who has the intention to pass through is photographed.

[0048] 3, in the case of a person to be verified who intends to pass through, the illuminance of the face image increases as time passes, as the person approaches the lighting installed in the ticket gate 40. Also, in the case of a person to be verified who intends to pass through, the person approaches the camera 10 installed in the ticket gate 40 as time passes, and the focus of the camera 10 is adjusted to the person to be verified, so the blur becomes smaller.

[0049] Fig. 4 is a diagram showing an example of the movement of a person to be verified who has no intention of passing through. Fig. 4 shows a time series of images taken by camera 10 installed in ticket gate 40 when the person to be verified who has no intention of passing through is photographed.

[0050] 4, in the case of a person to be verified who has no intention of passing through, even if time passes, the illuminance of the face image does not increase because the person does not approach the lighting mounted on the ticket gate 40. Also, in the case of a person to be verified who has no intention of passing through, even if time passes, the person does not approach the camera 10 provided on the ticket gate 40, and the focus of the camera 10 is not set on the person to be verified, so the blur does not decrease.

[0051] 3 and 4, the illuminance and blur differ depending on the presence or absence of an intention to pass. Therefore, in the present embodiment, for example, the presence or absence of an intention to pass is determined based on the illuminance and / or blur detected from a face image.

[0052] For example, the intention determination processing unit 202 of the control device 20 detects illuminance from the face image of the person to be matched, and if the illuminance is within a specified range, it determines that the person to be matched has an intention to pass, and if the illuminance is not within the specified range, it determines that the person to be matched has no intention to pass. Alternatively, the intention determination processing unit 202 detects blur from the face image of the person to be matched, and if the blur is within a specified range, it determines that the person to be matched has an intention to pass, and if the blur is not within the specified range, it determines that the person to be matched has no intention to pass. Alternatively, the intention determination processing unit 202 may detect illuminance and blur, and if the illuminance is within a specified range and / or the blur is within a specified range, it determines that the person to be matched has no intention to pass, and if not, it determines that the person to be matched has no intention to pass. As described above, the illuminance increases as it approaches the lighting mounted on the ticket gate 40, and the blur decreases as it approaches the camera 10 provided on the ticket gate 40. Therefore, the range of illuminance values ​​for determining that there is an intention to pass may be set to a range higher than a predetermined threshold, and the range of blur values ​​may be set to a range lower than a predetermined threshold.

[0053] The specified range to be compared with the illuminance and the specified range to be compared with the blur may be different. The specified range may be specified by both the upper and lower thresholds of the range, or one of the upper and lower thresholds of the range. For example, when the specified range to be compared with the illuminance is specified by only the lower threshold, the intention determination processing unit 202 determines that the person to be matched has an intention to pass if the illuminance is equal to or greater than the lower threshold, and determines that the person to be matched has no intention to pass if the illuminance is not equal to or greater than the lower threshold. The narrower the specified range (for example, the larger the lower threshold and / or the smaller the upper threshold), the stricter the criteria for determining that there is an intention to pass.

[0054] Also, for example, the intention determination processing unit 202 of the control device 20 may determine the presence or absence of an intention to pass based on a change in illuminance over time and / or a change in blur over time. For example, the intention determination processing unit 202 determines the presence or absence of an intention to pass based on a change between images of illuminance and / or blur detected from each of images captured at multiple time points.

[0055] For example, when determining whether or not there is an intention to pass based on a change in illuminance, the following determination is made. If the illuminance of the face image increases with time, the intention determination processing unit 202 determines that there is an intention to pass, and if the illuminance does not increase (i.e., if the illuminance does not change or decreases), the intention determination processing unit 202 determines that there is no intention to pass. When determining whether or not there is an intention to pass based on a change in blur, the following determination is made. If the blur of the face image decreases with time, the intention determination unit 202 determines that there is an intention to pass, and if the blur does not decrease (i.e., if the blur does not change or increases), the intention determination processing unit 202 determines that there is no intention to pass. Note that there is a risk that minute changes in both illuminance and blur may be detected due to subtle movements of the person and / or detection errors. If such minute changes are reflected in the determination of the intention to pass, the determination result becomes unstable, so changes within a predetermined range that are considered to be errors, etc. may not be used in the determination of the intention to pass. It should be noted that the change in illuminance and the change in blur here do not have to be calculated based on the illuminance and the blur detected from the images captured by the common camera 10.

[0056] The intention determination processing unit 202 of the control device 20 may determine the presence or absence of an intention to pass based on the value of one of the illuminance and blur and the change in the other. For example, it may be determined that there is an intention to pass when the illuminance is within a specified range and the change in blur matches a predetermined criterion. For example, in this case, the change in illuminance is not taken into account, but the change in blur is taken into account. In this way, by making a determination that does not take into account the change in either the illuminance or blur and takes into account the change in the other, it is possible to avoid the determination result becoming unstable due to a small change in either the illuminance or blur, and to improve the accuracy of the determination result based on both the illuminance and blur.

[0057] The specified range used to determine whether or not there is an intention to pass may be set in advance or may be dynamically changed based on the location of the ticket gate 40 (e.g., the brightness around the ticket gate 40, the flow of people around the ticket gate 40) and the performance of the camera 10 (focal length, exposure time, gain, etc.).

[0058] <Processing flow> Fig. 5 is a flow chart showing an example of the flow of face matching processing in the present embodiment. The flow shown in Fig. 5 is started, for example, when the operation of the ticket gate 40 is started.

[0059] The camera 10 captures an image of the area around the ticket gate 40 (S101).

[0060] The control device 20 performs face detection processing (S102). In the face detection processing, it is detected whether or not a face frame indicating an area corresponding to a person's face is present in the captured image.

[0061] The control device 20 determines whether or not a face frame has been detected (S103).

[0062] If a face frame is not detected (NO in S103), the control device 20 executes processing without a face frame (S104). In the processing without a face frame, for example, processing such as discarding the captured image may be performed. Alternatively, in the processing without a face frame, subsequent processing may not be performed. Then, the flow returns to S101.

[0063] When a face frame is detected (YES in S103), the control device 20 performs a passing intention determination process (S105). The passing intention determination process determines whether or not the person to be matched corresponding to the detected face frame has a passing intention. The flow of the passing intention determination process will be described later.

[0064] Based on the result of the passing intention determination process, the control device 20 determines whether or not the person to be verified has the intention to pass (S106).

[0065] If it is determined that there is no intention to pass (NO in S106), the control device 20 executes a process of no intention to pass (S107). The process of no intention to pass includes, for example, a process in which the control device 20 records that the detected face frame is a face frame with no intention to pass. Then, the flow returns to S101.

[0066] If it is determined that there is an intention to pass (YES in S106), the control device 20 requests the server device 1 to perform face matching processing (S108).

[0067] The control device 20 acquires a response to the request for face matching processing from the server device 1, and determines whether the person to be matched has the right to pass (S109).

[0068] If the person does not have the right of passage (NO in S109), the control device 20 determines that the person being verified may attempt unauthorized intrusion (S110).

[0069] If the person has the right to pass (YES in S109), the control device 20 determines that the person being verified has entered normally (S111).

[0070] Next, the passing intention determination process in S105 of FIG. 5 will be described.

[0071] FIG. 6 is a flowchart showing the flow of the passing intention determination process in S105 of FIG.

[0072] The control device 20 performs a face frame tracking process of the detected face frame (S201). In the face frame tracking process, the face frame detected at the current time point is compared with the face frame detected before the current time point, and the elapsed time, which is the difference in detection timing between the two face frames, and the coordinate change of the two face frames are determined. Then, based on the determined elapsed time and / or coordinate change, it is determined whether the two face frames are face frames of the same person to be matched. Note that whether the face frames are of the same person to be matched may be determined by other methods. For example, when the face matching process is performed in parallel with the passing intention determination process, it may be estimated whether the person to be matched is the same by referring to an intermediate result of the face matching process.

[0073] The control device 20 detects illuminance and blur from the face image of the currently detected face frame (S202).

[0074] The control device 20 determines whether the illuminance and blur are within a specified range (S203).

[0075] If at least one of the illuminance and the blur is not within the specified range (NO in S203), the process of S207 is executed.

[0076] If both the illuminance and the blur are within the specified range (YES in S203), the control device 20 judges whether the face frame detected at the current time point is a face frame that has already been detected before the current time point (S204). This judgment is made based on the result of the face frame tracking process in S201.

[0077] If the face frame detected currently is a face frame that has already been detected before the current time (YES at S204), the control device 20 determines whether or not the following conditions (hereinafter, "determination conditions for the difference") are satisfied: the difference between the illuminance of the face frame detected previously and the illuminance of the face frame detected this time (hereinafter, "illuminance difference") is within a specified range, and the difference between the blur of the face frame detected previously and the blur of the face frame detected this time (hereinafter, "blur difference") is within a specified range (S205).

[0078] If the face frame has not already been detected before the current time (NO in S204), or if the judgment condition for the difference is satisfied (YES in S205), the control device 20 judges that the person to be matched with the face frame detected at the current time has the intention to pass (S206).

[0079] If at least one of the illuminance and the blur is not within the specified range (NO in S203), or if the judgment condition for the difference is not satisfied (NO in S205), the control device 20 judges that the person being matched with the face frame currently detected has no intention of passing (S207).

[0080] Then, the flow shown in FIG. 6 ends, and the control device 20 executes the process of S106 in FIG.

[0081] In addition, in FIG. 6, an example in which the face frame tracking process is performed in S201 is shown, but the present disclosure is not limited thereto. For example, like the process of S205, when the determination based on the illuminance difference and / or the blur difference is not performed, the process of S201 may be omitted. Furthermore, when the process of S201 is omitted, the processes of S204 and S205 may be omitted. In this case, when both the illuminance and the blur are within the specified range in S203 (YES in S203), the control device 20 determines that the person to be matched whose face frame is currently detected has an intention to pass. Furthermore, when at least one of the illuminance and the blur is not within the specified range (NO in S203), the control device 20 determines that the person to be matched whose face frame is currently detected has no intention to pass (S207).

[0082] 6 shows an example in which illuminance and blur are used as parameters for determining whether a person wants to pass, but the present disclosure is not limited to this. One of illuminance and blur may be used as a parameter for determining whether a person wants to pass, and the other may not be used.

[0083] In addition, in FIG. 6, an example is shown in which illuminance and blur are used as parameters for determining whether or not a person wants to pass. However, the use of illuminance and the use of blur may be switched. Alternatively, one of illuminance and blur may be given more importance than the other. For example, in a case in which illuminance is given more importance than blur, if the determination result based on illuminance differs from the determination result based on blur, the determination result based on illuminance is adopted. In addition, if the determination result is obtained as a numerical value such as a probability reflecting the presence or absence of a will to pass (for example, the probability of passing through the gate within a predetermined time from the current time), the control device 20 may perform the determination of whether or not a person wants to pass by using a result obtained by combining the determination result based on illuminance and the determination result based on blur. In this case, the control device 20 may combine the two determination results based on illuminance and the determination result based on blur with different weights depending on whether or not the illuminance or blur is given more importance. In a case in which the determination result is obtained as a numerical value such as a probability, the control device 20 may determine that the person wants to pass if the numerical value is equal to or greater than a predetermined threshold, and may determine that the person does not want to pass if the numerical value is less than the predetermined threshold.

[0084] For example, if the surrounding light is too bright, the illuminance does not change much depending on the position of the person. Therefore, when the light around the ticket gate 40 is lower than a predetermined level (for example, in the evening, at night, on a cloudy day, etc.), the determination result based on the illuminance is prioritized over the determination result based on the blur. Specifically, out of the illuminance and the blur, only the illuminance is used to determine the intention to pass, or the two determination results are combined with a weighting that places more importance on the determination result based on the illuminance than on the determination result based on the blur. Note that, when the light around the ticket gate 40 is at or above a predetermined level (for example, during the day or on a sunny day), the determination result based on the blur is prioritized over the determination result based on the illuminance. Specifically, out of the illuminance and the blur, only the blur is used to determine the intention to pass, or the two determination results are combined with a weighting that places more importance on the determination result based on the blur than on the determination result based on the illuminance. Note that the method of detecting the brightness of the light around the ticket gate 40 is not particularly limited. For example, the control device 10 may detect the brightness of the surrounding light from the illuminance of an image captured by the camera 10, or may detect the brightness of the surrounding light using an image captured by a camera looking down on the ticket gate 40. Also, another device, instead of the control device 20, may detect the brightness of the surrounding light around the ticket gate 40, and the control device 20 may obtain the brightness of the surrounding light from the other device (a sensor that detects brightness).

[0085] For example, since blur reflects the difference between the focal distance and the distance of the person, the same result may be obtained when the person is close to the camera 10 and when the person is far from the camera 10. Therefore, when the distance between the person and the camera 10 is greater than the focal distance, the judgment result based on the blur is prioritized over the judgment result based on the illuminance. Specifically, out of the illuminance and the blur, only the blur is used to judge the intention to pass, or the two judgment results are combined with a weighting such that the judgment result based on the blur is given more importance than the judgment result based on the illuminance. Also, when the distance between the person and the camera 10 is closer than the focal distance, the judgment result based on the illuminance is prioritized over the judgment result based on the blur. Specifically, out of the illuminance and the blur, only the illuminance is used to judge the intention to pass, or the two judgment results are combined with a weighting such that the judgment result based on the illuminance is given more importance than the judgment result based on the blur. Note that the distance between the person and the camera 10 may be determined by a means other than the blur. For example, the control device 20 may use an image of a camera that captures an overhead view of the ticket gate 40 to determine the distance between the position of a part of the person that is thought to be the head and the camera 10. The control device 20 may also determine the distance between the person and the camera 10 from the size of the face frame. In this case, since the image of the person contains blur, the size of the face frame cannot be accurately estimated, and there is a possibility that the distance between the person and the camera 10 is erroneously determined. However, in this embodiment, the determination of the intention to pass is performed using illuminance or blur, so the distance between the person and the camera 10 only indirectly affects the determination of the intention to pass. Therefore, the error in the determination of the distance between the person and the camera 10 can be tolerated to a certain extent. Note that the distance between the person and the camera 10 may be determined by a device other than the control device 20, and the control device 20 may acquire the distance determined by the other device.

[0086] 6 shows an example in which the illuminance and blur of a face image are used as parameters for determining whether the person is willing to pass, but the present disclosure is not limited to this. The illuminance and blur of a part other than the face may be used as parameters for determining whether the person is willing to pass.

[0087] A parameter other than illuminance and blur may be used as a parameter for determining whether or not a person wants to pass. For example, at least one of the face direction, the change in coordinates of the face frame, and the change in size of the face frame may be used as a parameter for determining whether or not a person wants to pass. Alternatively, illuminance and / or blur may be used in combination with at least one of the face direction, the change in coordinates of the face frame, and the change in size of the face frame. Alternatively, a comprehensive determination may be made in response to an increase in information obtained in the future or a change in the camera position, without being limited to these parameters for determination.

[0088] Fig. 7 is a diagram showing examples of parameters used in determining whether or not a person has an intention to pass. Fig. 8 shows, in a table format, cases in which it is determined that there is no intention to pass for each parameter and cases in which it is determined that there is an intention to pass for each parameter.

[0089] 7 shows that, with regard to the face direction, when the face direction of the person to be verified faces left or right, it is determined that the person has no intention of passing through, and when the face direction faces directly ahead with respect to the camera 10, it is determined that the person has an intention of passing through. The face direction is an example of a parameter related to the posture of the person to be verified with respect to the ticket gate 40.

[0090] 7 shows that, with regard to illuminance, when the illuminance is below a predetermined level, it is determined that there is no intention to pass, and when the illuminance is equal to or higher than the predetermined level, it is determined that there is an intention to pass. Illuminance is an example of a parameter related to the position of the person to be verified relative to the ticket gate 40.

[0091] 7 shows that, with regard to the change in the coordinates of the face frame, if the coordinate in the Y-axis direction of the face frame does not change, it is determined that there is no intention to pass, and if the coordinate in the Y-axis direction of the face frame changes, it is determined that there is an intention to pass. Note that the Y-axis direction corresponds to the vertical direction in the captured image. In other words, if the position of the face frame in the captured image does not change in the vertical direction of the captured image, it is determined that there is no intention to pass, and if it changes in the vertical direction of the captured image, it is determined that there is an intention to pass. The change in the coordinates of the face frame is an example of a parameter related to the position of the person to be matched relative to the ticket gate 40.

[0092] 7 shows that, regarding blur, when the blur is at or above a predetermined level, it is determined that the person has no intention of passing through, and when the blur falls below the predetermined level, it is determined that the person has an intention of passing through. Blur is an example of a parameter related to the position of the person to be verified relative to the ticket gate 40.

[0093] 7 shows that, with regard to the size of the face frame, if there is no change in the size of the face frame or if the size of the face frame changes in a direction toward decreasing, it is determined that there is no intention to pass through, and if the size of the face frame changes in a direction toward increasing, it is determined that there is an intention to pass through. The size of the face frame is an example of a parameter related to the position of the person to be matched relative to the ticket gate 40.

[0094] As described above, in this embodiment, the communication unit 201 of the control device 20 acquires an image including a face image area, which is taken from the direction from the exit to the entrance of the ticket gate 40 of a person entering the ticket gate 40 (an example of a gate) where the passability is determined based on the result of the face matching process. Then, the intention determination processing unit 202 (an example of a control unit) determines whether or not to perform face matching processing of the person based on parameters (e.g., illuminance and / or blur) related to the posture of the person relative to the ticket gate 40 and / or the position of the person relative to the ticket gate 40 detected from the image. According to this configuration, it is determined whether or not a person included in the image captured by the camera 10 intends to pass through the ticket gate (in other words, whether or not to perform face matching), and the face matching process for the face image of a person who is determined to have no intention to pass through the ticket gate is omitted. This allows computer resources to be concentrated on the face matching process of a person who is determined to have the intention to pass through the ticket gate, the processing time of the face matching process in the ticket gate control can be reduced, and the speed of the face matching process can be increased. In addition, the targets of the face matching process can be reduced to those people who are determined to have the intention of passing through the ticket gate, thereby improving the accuracy of face matching.

[0095] According to the present embodiment, by judging the intention to pass based on illuminance and blur, for example, the judgment can be made without using parameters that vary depending on individuals, such as the size of the face of the person being verified, and the accuracy of judging the intention to pass can be improved. According to the present embodiment, the intention to pass can be judged using parameters that are less affected by items worn by the person being verified (e.g., hats, masks, glasses), and the accuracy of judging the intention to pass can be improved.

[0096] Moreover, according to this embodiment, the face matching process can be accelerated. The relationship between the face matching process and the process of determining whether or not a person has the intention to pass can be considered as follows: a process of determining whether or not a person has the intention to pass, and a process of performing face matching only on a person who has been determined to have the intention to pass; and a process of starting face matching independently of the process of determining whether or not a person has the intention to pass. In the former case, according to this embodiment, a person who has the intention to pass can be estimated before the person enters the ticket gate 40, so that the number of subjects to be subjected to face matching can be reduced, and the speed of the face matching process can be increased. In the latter case, according to this embodiment, the face matching process can be stopped at the time when it is determined that the person has no intention to pass, so that even if a person who has been determined to have no intention to pass is mistakenly set as a subject of the face matching process, the consumed computer resources can be suppressed. Therefore, regardless of whether or not a person has the intention to pass, the process before the face matching process and / or a part of the face matching process can be started at a stage when the person is away from the ticket gate 40, thereby increasing the speed of the face matching process. In particular, the results of the image capture and face frame detection processes, which are the pre-processing steps of the face matching process, can be used in both the face matching process and the process of determining the intention to pass. Therefore, by starting the pre-processing steps of the face matching process when the person is away from the ticket gate 40, the start of the face matching process after it is determined that there is an intention to pass can also be accelerated.

[0097] Furthermore, in this embodiment, image capture and face frame detection are performed as processes on the ticket gate side (e.g., camera 10 and control device 20), so if the processing related to the face matching process is stopped at the point of image capture and face frame detection, the control device 20 will not request face matching from the server device 30, thereby reducing the amount of communication between the ticket gate (e.g., control device 20) and the server device 30.

[0098] Furthermore, according to this embodiment, since the accuracy of the face matching process decreases when the ticket gate is located in a relatively dark place (for example, inside a subway station), even in cases where the face matching process cannot be started early, the processing time of the face matching process in ticket gate control can be reduced, thereby realizing faster face matching processing.

[0099] For example, in a case where the surrounding environment is relatively dark, the change in illuminance and the change in blur are larger than in a case where the surrounding environment is relatively bright. Therefore, the determination result of the intention to pass is less affected by an error in illuminance or blur, and this embodiment can improve the accuracy of the determination of the intention to pass, for example, in a case where the surrounding environment is dark, and can realize high-speed face matching processing. Note that a case where the surrounding environment is dark is, for example, a case such as a ticket gate at a subway station.

[0100] Also, a ticket gate using face matching processing is usually equipped with a camera and a light that illuminates a person to be matched who is about to pass through the ticket gate. This light may have a configuration in which light is irradiated from the exit of the ticket gate toward the entrance. For example, the light provided in the ticket gate may be a light that emits directional light such as a laser light, or may emit radial light such as a light bulb. Since the present embodiment uses a camera and light that are usually provided, when applying the system according to the present embodiment, no additional hardware is required. Therefore, the cost required for applying the system according to the present embodiment can be reduced. The light does not have to be configured as an integral part of the ticket gate. For example, a light source independent of the ticket gate may be installed separately, or may be configured as a window that receives natural light. In other words, the type and configuration of the light source are not important as long as the light is irradiated from the exit of the ticket gate toward the entrance.

[0101] In the above-described embodiment, multiple face frames may be detected in one captured image. When multiple face frames are detected, the face matching process including the above-described pass intention determination process is performed for each of the multiple face frames. By performing the above-described pass intention determination process for each of the multiple face frames, the priority order of the face matching process for the multiple face frames can be determined, so that the processing time of the face matching process in the ticket gate control can be reduced and the speed of the face matching process can be increased.

[0102] In the above-described embodiment, an example in which one camera 10 is provided has been shown, but the present disclosure is not limited thereto. A plurality of cameras 10 may be provided for the ticket gate 40. In this case, parameters used for determining the intention to pass may be detected from images captured by the plurality of cameras 10. Then, differences in parameters between the images captured by the plurality of cameras 10 may be calculated, and the presence or absence of the intention to pass may be determined based on the calculated differences in parameters.

[0103] In the above-described embodiment, an example was shown in which the illuminance and blur of a face image are used as parameters for determining whether or not a person wants to pass, but the present disclosure is not limited to this. The illuminance and blur of a part other than the face may be used as parameters for determining whether or not a person wants to pass. Furthermore, the illuminance and blur of a part other than the face may be used in combination with the illuminance and blur of the face image.

[0104] The illuminance and blur of the portion other than the face of a person included in one captured image and the illuminance and blur of the face image are consistent. For example, the difference between the illuminance of the portion other than the face of a person and the illuminance of the face image is less than a predetermined value. Also, for example, the difference between the blur of the portion other than the face of a person and the blur of the face image is less than a predetermined value.

[0105] On the other hand, it is considered that the illuminance and blur of the parts other than the faces of different people included in one photographed image may not be consistent with the illuminance and blur of the face image. For example, if two people included in one photographed image are person #1 and person #2, the difference in illuminance between the parts other than the face of person #2 and the illuminance of the face image of person #1 is likely to be a predetermined value or more.

[0106] Therefore, whether or not to cut in line at the ticket gate may be determined based on the relationship between the illuminance and blur of parts other than the face and the illuminance and blur of the face image.

[0107] Fig. 8 is a diagram showing an example of cut-in determination. Fig. 8 shows person #1, whose face frame has been detected, and person #2, who cuts in front of person #1, in an image captured by camera 10. Note that person #2 intentionally hides his face, so the face frame is not detected.

[0108] In Fig. 8, the distance from camera 10 differs between the two persons, so the illuminance of the face image of person #1 is darker than the illuminance of the portion other than the face of person #2 (upper body in Fig. 8). Therefore, the difference in illuminance between the face image of person #1's face frame and the illuminance of the portion other than the face of person #2 may be equal to or greater than a predetermined value. In this way, by comparing the difference in illuminance with a predetermined value, it is possible to determine whether an interruption has occurred.

[0109] In the above-described embodiment, the method of the passing intention determination process and / or parameters related to the passing intention determination process may be adjusted according to the surrounding environment.

[0110] For example, when the surroundings of the ticket gate 40 are brighter than a predetermined value, the intensity of the light irradiated onto the range photographed by the camera 10 is adjusted.

[0111] For example, if the surroundings of the ticket gate 40 are brighter than a predetermined value due to artificial lighting, the lighting near the ticket gate 40 is set to a lower value since the surrounding brightness is stable.

[0112] For example, if the surroundings of the ticket gate 40 are brighter than a predetermined value due to natural light, the influence of the lighting of the ticket gate 40 becomes greater the closer one gets to the ticket gate 40. Therefore, the criteria for determining that "there is an intention to pass" are adjusted as follows. When natural light is backlit relative to the camera 10, for example, the ticket gate control unit 204 of the control device 20 sets the lighting of the ticket gate 40 to be brighter. This increases the illuminance of the person as they approach the ticket gate 40. Using this property, the intention determination processing unit 202 of the control device 20 determines that the person has an intention to pass if it detects that the illuminance of the person has increased, and determines that the person has no intention to pass if it detects that the illuminance of the person has not increased. When natural light is direct light to the camera 10, the ticket gate control unit 204 of the control device 20 sets the lighting of the ticket gate 40 to be darker. This reduces the illuminance of a person as they approach the ticket gate 40. Using this property, the intention determination processing unit 202 of the control device 20 determines that the person has an intention to pass when it detects that the illuminance of the person has decreased, and determines that the person has no intention to pass when it detects that the illuminance of the person has not decreased.

[0113] In this case, backlighting refers to light shining from behind the person approaching the ticket gate 40 (or camera 10), and frontlighting refers to light shining from the direction of camera 10 in the direction of the person approaching the ticket gate 40 (or camera 10).

[0114] Backlighting and / or frontlighting may be automatically determined based on the brightness of reference points set in the image, and the criteria for determining whether or not there is an intention to pass and the brightness of the lighting may be adjusted based on the determination result.

[0115] The evaluation result of the accuracy of the passing intention determination may be corrected based on the criteria for determining whether the vehicle is in front of or behind the light. For example, if an accurate result is obtained with the settings for front light, the criteria for determining whether the vehicle is in front of the light may be set to be stricter in the case of back light.

[0116] In the above example, the brightness of the lighting in the ticket gate 40 is adjusted depending on whether backlighting or frontlighting is occurring, but the adjustment of the lighting may be omitted. However, since adjusting the brightness of the lighting has the effect of making the effects of backlighting or frontlighting more prominent, adjusting the brightness of the lighting will improve the accuracy of determining the intention to pass.

[0117] In addition, the light source that generates backlight or frontlight is not limited to natural light, and backlight or frontlight may occur due to the influence of other light sources, such as the light source of the facility in which the ticket gate 40 is installed. In these cases, the presence or absence of the intention to pass may be determined using the same criteria as above.

[0118] In the above embodiment, the face matching process is performed after the determination of the intention to pass, but the face matching process may be started regardless of whether the determination of the intention to pass is completed. In this case, the processing load can be reduced by stopping the face matching process as soon as it is determined that there is no intention to pass.

[0119] For example, if a person enters inside the ticket gate 40, face matching processing is forcibly started. The limit position at which face matching processing starts is described as the face matching start limit position. The face matching start limit position is a position at which it is assumed that the person will not pass through the ticket gate 40 before face matching is completed. If face matching starts at a position closer to the ticket gate 40 than the face matching start limit position, the person will attempt to pass through the ticket gate 40 before face matching is completed. Note that the determination of whether or not a person has passed through the face matching start limit position may be made using a method other than that for determining the intention to pass (for example, a method using a sensor).

[0120] If the person advances to the face matching start limit position without completing the determination of the intention to pass, there is a high possibility that the determination of the intention to pass has failed. Therefore, if the person advances to the face matching start limit position, the control device 20 may carry out the face matching process to the end even if the determination result of the intention to pass is obtained thereafter.

[0121] Also, the control device 20 may start a pre-stage process of the face matching process and / or a part of the face matching process from a position (face matching start position) before the face matching start limit position. In this case, the pre-stage process of the face matching process and / or a part of the face matching process is started at the stage when the person enters the face matching start position. For example, image capture and face frame detection are started as the pre-stage process of the face matching process. The result of the image capture and face frame detection process can be used to determine whether or not the person intends to pass, while the process after the image capture and face frame detection process is likely to be a wasteful process if the person does not intend to pass. Therefore, the process after the image capture and face frame detection process may be started only when the person enters the face matching start limit position or when it is determined that the person intends to pass. The face matching start position may be adjusted according to the brightness of the surroundings. For example, the face matching start position may be brought closer to the ticket gate 40 because the brighter the surroundings are, the more accurate and faster the face matching can be performed.

[0122] Fig. 9 is a diagram showing an example of the relationship between the face matching start position and brightness. Fig. 9 shows three cases when the ticket gate 40 is viewed from above. The "bright case" in Fig. 9 corresponds to a case where the surrounding brightness is equal to or higher than a predetermined level, and the "dark case" in Fig. 9 corresponds to a case where the surrounding brightness is less than the predetermined level. In each of the three cases, the left side shows a state where a person is heading towards the ticket gate 40, that is, a state where there is an intention to pass through, and the right side shows a state where a person is not heading towards the ticket gate 40, that is, a state where there is no intention to pass through.

[0123] Cases 1 and 2 in FIG. 9 are "dark cases." In the "dark case," the face matching start position is set farther from the ticket gate 40 than in the "light case." In this case, the control device 20 performs image capture and face frame detection processing on persons who have passed the face matching start position, and determines the intention to pass based on the result, thereby making it possible to exclude persons who have no intention to pass (person on the right in case 2) from the targets of face matching processing. This reduces the possibility of unnecessary face matching processing being performed. In other words, the determination of the intention to pass in this embodiment has the effect of reducing the processing load, particularly in the "dark case."

[0124] On the other hand, in the "bright case", since the face matching process can be completed in a short time, it is possible to exclude people who have no intention of passing so that they are not the target of the face matching process in the first place by moving the face matching start position closer to the ticket gate 40. In such a case, even if the process of determining whether or not to pass is performed, it is unlikely that a result will be obtained before the person reaches the face matching start limit position, so the determination of whether or not to pass may not be performed.

[0125] Also, the face matching start position may be the position where the face matching process is started. That is, the image capturing and face frame detection processes may be completed before the person reaches the face matching start position. In this way, the control device 20 can determine whether or not the person has reached the face matching start position using the result of the face frame detection, so that other sensors for estimating the position of the person (for example, the distance between the person and the camera 10) are not required. In this case, the face matching process may be performed only on the person who is determined to have the intention to pass based on the result of the image capturing and face frame detection. In this way, the person who is the subject of the face matching process with a heavy load can be limited to the person who is determined to have the intention to pass. Furthermore, in this case, the face matching start position may be adjustable by adjusting the criteria for determining that the person has the intention to pass. For example, as shown in FIG. 9, the brighter the surroundings are, the stricter the criteria for determining that the person has the intention to pass may be, and the face matching start position may be closer to the ticket gate 40. In this case, as shown in Case 2, the face matching process of the person who has the intention to pass is performed at a position closer to the ticket gate 40 than the face matching start position. When the criteria for determining that there is an intention to pass are made stricter, the criteria for determining that there is an intention to pass may be, for example, a larger illuminance value, a smaller blur value, or a larger change in illuminance or blur. For example, in a case where the intention determination processing unit 202 of the control device 20 determines that the person to be matched has an intention to pass when the illuminance is equal to or greater than a threshold, the criteria for determining that there is an intention to pass may be made stricter by increasing the threshold value compared with the illuminance. In this way, it becomes more difficult to determine that there is an intention to pass unless the person is closer to the ticket gate 40.

[0126] When the criteria for determining that there is an intention to pass are made stricter, the position where the face matching process starts (face matching start position) moves closer to the ticket gate 40. In such a case, the face recognition start position can be moved closer to the ticket gate 40 by moving the focus of the camera 10 closer to the ticket gate 40 or shortening the exposure time. Note that when making these adjustments, the brightness of the face in the captured image also changes, so control may be performed to emphasize the determination result of the intention to pass due to blur.

[0127] A process for avoiding errors in the control of the ticket gate, including errors in the face matching process, may be added. For example, the accuracy of the determination of the intention to pass may be evaluated based on the rate at which a person whose face matching has started near the authentication start position actually passes through the ticket gate 40. For example, if the rate at which a person whose face matching has started passes through the ticket gate 40 is lower than a threshold value, an alert may be issued to prompt the user to readjust the ticket gate 40.

[0128] In the above-mentioned embodiment, a ticket gate at a station is taken as an example for explanation, but the present disclosure is not limited to a ticket gate at a station. The present disclosure may be applied to a gate installed at an airport, an office, or the like other than a station, which allows entry of specific persons and blocks entry of persons other than the specific persons. The present disclosure may also be applied to a gate that does not have a door that restricts the passage of persons.

[0129] In the above-described embodiment, when a configuration for determining the intention to pass based on a change in illuminance or blur is adopted, if there is no change in illuminance or blur, it may be determined that there is an intention to pass, or that there is no intention to pass. Such a situation occurs, for example, when a person stops. For example, if there is an intention to pass, a person may stop to leave a gap between him / herself and a person ahead, or for other reasons. Also, for example, if there is no intention to pass, a person may stop to observe a ticket gate, or for other reasons. Therefore, if there is no change in illuminance or blur that is expected to cause a stop, it may be determined that there is an intention to pass, or that there is no intention to pass, depending on the intention of the operator who manages the passage management system including the ticket gate. Also, a determination result indicating a stop may be provided, and it may be determined that there is an intention to pass or that there is no intention to pass. Also, a determination result indicating a stop may be provided, and one of three determination results may be obtained: a determination result that there is an intention to pass, a determination result that there is no intention to pass, and a determination result that there is a stop. In this case, for a person determined to be standing still, additional information may be used to determine whether to treat the person as having an intention to pass or not. For example, if there is a margin in the current overall system processing capacity (for example, if the load of face matching processing is less than a predetermined level), the person may be treated as having an intention to pass even when he or she is standing still, and if there is no margin in the current overall system processing capacity (for example, if the load of face matching processing is equal to or greater than a predetermined level), the person may be treated as having no intention to pass. Note that the absence of change in illuminance or blur may be considered to mean a case where the change in illuminance or blur is within a predetermined range.

[0130] The present disclosure may be applied to a gateless passage management system. For example, in a case where a movement from a first point through a second point to a third point is permitted for a specific person and not permitted for people other than the specific person, whether or not a certain person intends to move to the third point may be determined based on a parameter detected from an image taken from the second point to the first point.

[0131] The present disclosure can be realized in software, hardware, or software in conjunction with hardware.

[0132] Each functional block used in the description of the above embodiments may be partially or entirely realized as an LSI, which is an integrated circuit, and each process described in the above embodiments may be partially or entirely controlled by one LSI or a combination of LSIs. The LSI may be composed of individual chips, or may be composed of one chip that includes some or all of the functional blocks. The LSI may have data input and output. Depending on the degree of integration, the LSI may be called an IC, a system LSI, a super LSI, or an ultra LSI.

[0133] The method of integration is not limited to LSI, and may be realized by a dedicated circuit, a general-purpose processor, or a dedicated processor. Also, after LSI manufacturing, a programmable FPGA (Field Programmable Gate Array) or a reconfigurable processor that can reconfigure the connections and settings of circuit cells inside the LSI may be used. The present disclosure may be realized as digital processing or analog processing.

[0134] Furthermore, if a new integrated circuit technology that can replace LSI appears due to the progress of semiconductor technology or a derivative technology, it is possible to integrate the functional blocks using that technology. The application of biotechnology is also a possibility.

[0135] The present disclosure may be implemented in any type of apparatus, device, or system having a communication function (collectively referred to as a communication apparatus). The communication apparatus may include a radio transceiver and processing / control circuitry. The radio transceiver may include a receiver and a transmitter, or both as functions. The radio transceiver (transmitter and receiver) may include a radio frequency (RF) module and one or more antennas. The RF module may include an amplifier, an RF modulator / demodulator, or the like. Non-limiting examples of communication devices include telephones (e.g., cell phones, smartphones, etc.), tablets, personal computers (PCs) (e.g., laptops, desktops, notebooks, etc.), cameras (e.g., digital still / video cameras), digital players (e.g., digital audio / video players), wearable devices (e.g., wearable cameras, smart watches, tracking devices, etc.), game consoles, digital book readers, telehealth and telemedicine devices, communication-enabled vehicles or mobile conveyances (e.g., cars, airplanes, boats, etc.), and combinations of the above devices.

[0136] Communications Equipment includes, but is not limited to portable or mobile equipment, non-portable or fixed equipment, devices and systems of any kind, such as smart home devices (appliances, lighting equipment, smart meters or metering devices, control panels, etc.), vending machines and any other "Things" that may exist on an Internet of Things (IoT) network.

[0137] In recent years, in the field of IoT (Internet of Things) technology, CPS (Cyber ​​Physical Systems) is a new concept that creates new added value by linking information between physical space and cyber space. This CPS concept can also be adopted in the above-mentioned embodiment.

[0138] That is, as a basic configuration of a CPS, for example, an edge server located in a physical space and a cloud server located in a cyberspace can be connected via a network, and processing can be distributed and processed by the processors installed in both servers. Here, each processing data generated in the edge server or the cloud server is preferably generated on a standardized platform, and the use of such a standardized platform can improve the efficiency of building a system including a variety of sensors and IoT application software.

[0139] Communications include data communications via cellular systems, wireless LAN systems, communications satellite systems, etc., as well as data communications via combinations of these.

[0140] A communications apparatus also includes devices, such as controllers and sensors, connected or coupled to a communications device that performs the communications functions described in this disclosure, such as controllers and sensors that generate control and data signals used by the communications device to perform the communications functions of the communications apparatus.

[0141] The communication apparatus also includes infrastructure facilities, such as base stations, access points, and any other apparatus, device, or system that communicates with or controls the various apparatuses listed above, but are not limited to these.

[0142] 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 modified or amended examples within the scope of the claims, and it is understood that these also naturally belong to the technical scope of the present disclosure. In addition, the components in the above-mentioned embodiments may be arbitrarily combined within the scope of the disclosure.

[0143] Although specific examples of the present disclosure have been described in detail above, these are merely examples and do not limit the scope of the claims. The technology described in the claims includes various modifications and alterations of the specific examples exemplified above. [Industrial Applicability]

[0144] An embodiment of the present disclosure is suitable for a face recognition system. [Explanation of symbols]

[0145] 1. Currency Management System 10. Camera 20 Control device 30 Server device 40 Ticket Gates 201 Communications Department 202 Intention determination processing unit 203 Face matching processing unit 204 Ticket gate control unit 301 Storage section 302 Face matching unit 303 Communications Department

Claims

1. an acquisition unit that acquires an image including a face image area of ​​a person entering a gate, the person being allowed to pass through which is determined based on a result of face matching processing, taken from a direction from an exit of the gate toward an entrance of the gate; a control unit that determines whether or not to perform the face matching process on the person based on illuminance and / or blur of the person detected from the image; A control device comprising:

2. the person is illuminated by a light source that emits light in a direction from the exit of the gate to the entrance of the gate, the control unit determines whether or not to perform the face matching process of the person based on a change in the illuminance detected from each of the images captured at a plurality of time points between the images. The control device according to claim 1 .

3. The control unit is determining that a face matching process for the person is to be performed when the illuminance of the person detected from the image captured at a later time point is higher than the illuminance of the person detected from the image captured at an earlier time point; determining that a face matching process for the person is not to be performed when an illuminance of the person detected from the image captured at the later time point is lower than an illuminance of the person detected from the image captured at the previous time point; The control device according to claim 2.

4. The light source is a light provided at the gate, the person is further illuminated by ambient light around the gate; When the surrounding light is backlighting the person, the control unit determining that a face matching process for the person is to be performed when the illuminance of the person detected from the image captured at a later time point is higher than the illuminance of the person detected from the image captured at an earlier time point; determining that a face matching process for the person is not to be performed when an illuminance of the person detected from the image captured at the later time point is lower than an illuminance of the person detected from the image captured at the previous time point; The control device according to claim 2.

5. The light source is a light provided at the gate, The control unit controls the brightness of the illumination to be higher when the surrounding light is backlighting the person than when the surrounding light is not backlighting the person. The control device according to claim 2.

6. The light source is a light provided at the gate, the person is further illuminated by ambient light around the gate; When the surrounding light is direct light with respect to the person, the control unit determining that a face matching process for the person is to be performed when the illuminance of the person detected from the image captured at the later time point is lower than the illuminance of the person detected from the image captured at the earlier time point; determining that a face matching process for the person is not to be performed when an illuminance of the person detected from the image captured at the later time point is higher than an illuminance of the person detected from the image captured at the previous time point; The control device according to claim 2.

7. The light source is a light provided at the gate, The control unit controls the brightness of the illumination to be lower when the surrounding light is a front light with respect to the person than when the surrounding light is not a front light with respect to the person. The control device according to claim 2.

8. the control unit determines whether to perform the face matching process of the person based on the illuminance and the blur of the person; the control unit performs control such that, when the brightness around the gate is equal to or higher than a predetermined level, a result of the determination based on the blur is given priority over a result of the determination based on the illuminance in the determination. The control device according to claim 2.

9. The control unit further acquires a distance between the person and a camera that captured an image including the face image area, the control unit determines whether to perform the face matching process of the person based on the illuminance and the blur of the person; When the distance is shorter than a predetermined threshold, the control unit performs control so that a result of the determination based on the blur is given priority over a result of the determination based on the illuminance in the determination. The control device according to claim 2.

10. When the control unit determines that the face matching process of the person is to be performed, the control unit performs the face matching process of identifying the person using the face image area, and when the control unit determines that the face matching process of the person is not to be performed, the control unit omits the face matching process of the person. The control device according to claim 1 .

11. The control unit further determines whether or not to perform the face matching process of the person based on at least one of a position, a size, and a face direction of the person of the face image area included in the image. The control device according to claim 1 .

12. a reference position for determining whether or not to perform the face matching process of the person when the brightness of the surroundings of the gate is equal to or higher than a predetermined level is closer to the gate than a reference position for determining whether or not to perform the face matching process of the person when the brightness of the surroundings of the gate is less than the predetermined level; The control device according to claim 1 .

13. The control device, A person entering a gate, the passability of which is determined based on the result of a face matching process, is photographed from a direction from the exit of the gate toward the entrance, and an image including a face image area is obtained; determining whether to perform the face matching process on the person based on the illuminance and / or blur of the person detected from the image; Control methods.

Images