Vehicle detection device, vehicle detection method, and vehicle detection program

Abstract

To propose a vehicle detection device, a vehicle detection method, and a vehicle detection program that can reduce processing load while maintaining detection accuracy.SOLUTION: A vehicle detection device includes a detection unit and an acquisition unit. The detection unit performs detection processing to detect vehicle numbers of a vehicle passing through a predetermined area at predetermined intervals from images continuously captured in accordance with an imaging cycle of a camera. The acquisition unit acquires vehicle information indicating the vehicle when the vehicle passes through the specified area. The detection unit performs detection processing at a first cycle longer than the imaging cycle when the acquisition unit has not acquired the vehicle information, and performs detection processing at a second cycle shorter than the first cycle when the acquisition unit has acquired the vehicle information.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present disclosure relates to a vehicle detection device, a vehicle detection method, and a vehicle detection program. [Background technology]

[0002] Conventionally, there is a technology for managing entry and exit of a parking lot by detecting vehicles passing through the entrance and exit of the parking lot, etc. This type of technology is known to detect vehicles by recognizing the numbers and letters on license plates from images captured by a camera (for example, see Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 2595510 Summary of the Invention [Problem to be solved by the invention]

[0004] In the prior art, the accuracy of license plate detection can be ensured by shortening the processing cycle of the detection process, such as by performing the detection process each time an image is captured. However, shortening the processing cycle increases the processing load.

[0005] Therefore, the present disclosure proposes a vehicle detection device, a vehicle detection method, and a vehicle detection program that can reduce the processing load while maintaining detection accuracy. [Means for solving the problem]

[0006] In order to solve the above problem, the vehicle detection device according to the present disclosure includes a detection unit and an acquisition unit. The detection unit performs a detection process to detect the vehicle license plate number of a vehicle passing through a predetermined area at a predetermined cycle from images captured continuously according to a camera's imaging cycle. The acquisition unit acquires vehicle information indicating the vehicle when the vehicle passes through the predetermined area. If the acquisition unit has not acquired the vehicle information, the detection unit performs the detection process at a first cycle longer than the imaging cycle, and if the acquisition unit has acquired the vehicle information, the detection unit performs the detection process at a second cycle shorter than the first cycle.

[0007] This allows the vehicle detection device to reduce the processing load while maintaining detection accuracy.

[0008] The acquisition unit according to the present disclosure acquires a vehicle number as the vehicle information, and the detection unit performs the detection process at least one of before and after the acquisition timing of the vehicle number for each second period.

[0009] This allows the vehicle detection device to reduce the processing load while maintaining detection accuracy.

[0010] The acquisition unit according to the present disclosure sets, in the image, a first region corresponding to the near side in the traveling direction of the vehicle and a second region corresponding to the far side in the traveling direction, and determines which region the vehicle number is located in. If the vehicle number is present in the first region, the detection unit performs the detection process for the images obtained after the acquisition timing at the second cycle, and if the vehicle number is present in the second region, the detection process for the images obtained before the acquisition timing at the second cycle.

[0011] This allows the vehicle detection device to reduce the processing load while maintaining detection accuracy.

[0012] In addition, the detection unit of the present disclosure, when the vehicle number is present in the first area, performs the detection process on images after the acquisition timing every second period, and then performs the detection process on images before the acquisition timing every second period; when the vehicle number is present in the second area, performs the detection process on images before the acquisition timing every second period, and then performs the detection process on images after the acquisition timing every second period.

[0013] This allows the vehicle detection device to improve the detection accuracy of vehicle license plates.

[0014] Furthermore, the acquisition unit according to the present disclosure sets a third area between the first area and the second area, and determines which area the vehicle number is located in. If the vehicle number is in the third area, the detection unit performs the detection process alternately for each of the images taken after the acquisition timing and the images taken before the acquisition timing, every second period.

[0015] This allows the vehicle detection device to reduce the processing load while maintaining detection accuracy.

[0016] The present disclosure further includes a storage unit that stores a most recent vehicle number confirmation list, which is a list of vehicle numbers detected by the detection unit during a period in which the vehicle passes through the specified area and confirmed as detection results. If the vehicle number acquired by the acquisition unit matches the vehicle number included in the most recent vehicle number confirmation list, the detection unit does not perform the detection process for each second period and confirms the vehicle number acquired by the acquisition unit as a detection result.

[0017] This allows the vehicle detection device to reduce the processing load.

[0018] In addition, when the vehicle number acquired by the acquisition unit matches a pre-registered vehicle number, the detection unit according to the present disclosure does not perform the detection process for each second period, and determines the vehicle number acquired by the acquisition unit as the detection result.

[0019] This allows the vehicle detection device to reduce the processing load.

[0020] Furthermore, the second period according to the present disclosure is the same period as the imaging period.

[0021] This allows the vehicle detection device to reduce the processing load while maintaining detection accuracy.

[0022] Furthermore, the detection section according to the present disclosure determines a target period for the detection process to be performed in each second cycle based on environmental information indicating the environment of the predetermined area.

[0023] This allows the vehicle detection device to improve the detection accuracy of vehicle license plates.

[0024] The vehicle detection method according to the present disclosure is a computer-executed vehicle detection method that includes a detection step and an acquisition step. The detection step performs a detection process that detects the vehicle license plate number of a vehicle passing through a predetermined area at a predetermined interval from images captured continuously according to a camera's imaging cycle. The acquisition step acquires vehicle information indicating the vehicle when the vehicle passes through the predetermined area. If the acquisition step has not acquired the vehicle information, the detection step performs the detection process at a first interval that is longer than the imaging cycle, and if the acquisition step has acquired the vehicle information, the detection process performs the detection process at a second interval that is shorter than the first interval.

[0025] This allows the vehicle detection method to reduce the processing load while maintaining detection accuracy.

[0026] A vehicle detection program according to the present disclosure is a vehicle detection program executed by a computer, and causes the computer to execute a detection procedure and an acquisition procedure. The detection procedure performs a detection process of detecting the vehicle license plate number of a vehicle passing through a predetermined area at a predetermined cycle from images continuously captured according to a camera's imaging cycle. The acquisition procedure acquires vehicle information indicating the vehicle when the vehicle passes through the predetermined area. If the acquisition procedure has not acquired the vehicle information, the detection procedure performs the detection process at a first cycle longer than the imaging cycle, and if the acquisition procedure has acquired the vehicle information, the detection process at a second cycle shorter than the first cycle.

[0027] This allows the vehicle detection program to reduce the processing load while maintaining detection accuracy. [Brief explanation of the drawings]

[0028] [Figure 1A] 1 is a diagram illustrating an example of the configuration of an entry / exit management system according to an embodiment; [Figure 1B] FIG. 1 is a diagram illustrating an overview of a vehicle detection method according to an embodiment. [Figure 2] FIG. 2 is a functional block diagram illustrating an example of the configuration of a management device according to the embodiment. [Figure 3] FIG. 10 is a diagram illustrating an example of a registration information DB. [Figure 4] FIG. 4 is a diagram illustrating an example of travel history information. [Figure 5] FIG. 10 is a diagram illustrating an example of a detection process performed by a detection unit. [Figure 6] 10A and 10B are diagrams illustrating another example of the detection process performed by the detection unit. [Figure 7] FIG. 10 is a diagram showing an example of a screen provided by a providing unit. [Figure 8] FIG. 10 is a diagram showing an example of a screen displayed on a video screen. [Figure 9] 10 is a flowchart illustrating the procedure of an overall process executed by a management device according to an embodiment. [Figure 10] 10 is a flowchart showing the procedure of a vehicle number detection process executed by the management device according to the embodiment. [Figure 11] 10 is a flowchart showing the procedure of a vehicle number determination process executed by the management device according to the embodiment. [Figure 12] FIG. 10 is a diagram showing an example of a most recent vehicle number confirmation list. [Figure 13] FIG. 10 is a diagram showing an example of a vehicle number detection list. DETAILED DESCRIPTION OF THE INVENTION

[0029] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. In the following embodiments, the same components are designated by the same reference numerals, and redundant description will be omitted.

[0030] First, an overview of a vehicle detection method according to an embodiment will be described with reference to Figures 1A and 1B. Figure 1A is a diagram illustrating an example of the configuration of an access control system according to an embodiment. Figure 1B is a diagram illustrating an overview of a vehicle detection method according to an embodiment.

[0031] As shown in FIG. 1A, the access control system S is installed in an area R where vehicles can enter. The vehicles are, for example, various types of vehicles such as cars, buses, trucks, and motorcycles. Area R is an area where vehicles can move within area R, such as a parking lot or a company premises. Area R is also provided with an entrance / exit 110, and vehicles enter or exit area R by passing through entrance / exit 110.

[0032] 1A shows an example in which entrance / exit 110 is a road that is shared by vehicles entering area R and vehicles exiting area R, but the entrance and exit may be provided separately. Also, entrance / exit 110 may be provided with an entrance / exit gate that opens and closes when vehicles enter and exit the area.

[0033] The entry / exit management system S according to the embodiment manages the entry and exit of vehicles in an area R based on information acquired from vehicles passing through the entrance / exit 110. Specifically, the entry / exit management system S according to the embodiment manages the entry and exit by generating a traffic history that associates the vehicle number of the vehicle (the number information shown on the license plate, hereinafter also referred to as the vehicle number) detected when the vehicle passes through the entrance / exit 110 with the identification information of the on-board device installed in the vehicle.

[0034] Here, the configuration of the entrance / exit control system S will be described. As shown in Fig. 1A, the entrance / exit control system S includes a management device 1 (vehicle detection device), a roadside device 10, an acquisition device 11, an imaging device 12, and a server device 200. Note that, although the present disclosure has exemplified a case where the vehicle detection device is the management device 1, a vehicle detection device may be provided separately from the management device 1.

[0035] The acquisition device 11 is placed, for example, above the entrance / exit 110, and communicates with an on-board device mounted on a vehicle located within the communication range of the communication antenna to acquire identification information of the on-board device. Specifically, the acquisition device 11 acquires the identification information multiple times at predetermined intervals while the vehicle is located within the communication range.

[0036] The vehicle-mounted device is, for example, an ETC (Electronic Toll Collection System) device, and transmits a WCN (Wireless Call Number) as identification information to the acquisition device 11. The acquisition device 11 transmits the identification information acquired from the vehicle-mounted device to the roadside device 10.

[0037] 1A shows an example in which the acquisition device 11 is placed on the entrance side of the entrance / exit 110 and communicates with vehicles entering the area R, but the acquisition device 11 may be placed on the exit side of the entrance / exit 110 and communicate with vehicles exiting the area R. Alternatively, the acquisition device 11 may be placed on both the entrance side and the exit side of the entrance / exit 110.

[0038] The imaging device 12 is a camera whose imaging range is the periphery of the entrance / exit 110. For example, the imaging device 12 is placed at a predetermined height relative to the roadside unit 10. The predetermined height is, for example, a height corresponding to the height of a license plate (an example of a vehicle number) of a vehicle. In other words, the imaging device 12 captures an image of an area including the license plate of the vehicle. The imaging device 12 transmits the captured vehicle image to the roadside unit 10.

[0039] 1A shows an example in which imaging device 12 is placed on the entrance side of entrance / exit 110 and captures an image of a vehicle entering area R, but imaging device 12 may also be placed on the exit side of entrance / exit 110 and capture an image of a vehicle exiting area R. Alternatively, imaging device 12 may be placed on both the entrance side and the exit side of entrance / exit 110.

[0040] The roadside unit 10 is a pole-shaped device placed beside the entrance / exit 110. The roadside unit 10 is communicably connected to the management device 1 and transmits the identification information acquired from the acquisition device 11 and the vehicle image acquired from the imaging device 12 to the management device 1.

[0041] The server device 200 is a server device that is communicably connected to the management device 1 wirelessly or via a wired connection. When registering a vehicle that enters or leaves the area R, the server device 200 registers and stores the vehicle information (vehicle number and identification information) acquired from the management device 1 in the registration information DB 210. Details of the registration information DB 210 will be described later with reference to FIG. 3.

[0042] The management device 1 is a terminal device operated by an administrator who manages the access control system S. For example, the management device 1 can be any type of terminal device, such as a smartphone, a desktop PC, a notebook PC, or a tablet PC.

[0043] The management device 1 acquires identification information and vehicle images from the roadside device 10. Specifically, the management device 1 acquires vehicle images captured continuously at predetermined intervals by the imaging device 12. Furthermore, the management device 1 detects the numbers and letters (together, the vehicle number) on the license plate of the vehicle based on the acquired vehicle images.

[0044] The management device 1 then associates the acquired identification information with the vehicle number, generates a travel history including the associated vehicle number and identification information, and stores the generated travel history information 32.

[0045] In the present disclosure, the processing load of the management device 1 is reduced by devising a processing interval for a detection process for detecting a vehicle number plate from a vehicle image. Here, the processing interval for the detection process according to the present disclosure will be explained using FIG. 1B. In FIG. 1B, it is assumed that an image is captured at each of times t1 to t12. Also, in FIG. 1B, it is assumed that the vehicle number plate is first detected (acquired) in the image at time t9.

[0046] In the vehicle detection method of the embodiment, if the vehicle number is not acquired, a detection process is performed to detect the vehicle number at a first period P1 that is longer than the imaging period (period P2), and if the vehicle number is acquired, a detection process is performed at a second period P2 that is shorter than the first period P1.

[0047] For example, in FIG. 1B, the management device 1 performs the detection process at the timing of a first cycle P1 during the period from time t1 to time t9 when the vehicle number has not been acquired. That is, in FIG. 1B, the management device 1 performs the detection process using images captured at time t1, time t5, and time t9. Then, when the management device 1 detects (acquires) a vehicle number at time t9, it performs the detection process using images captured in the periods before and after time t9. Specifically, when the management device 1 detects a vehicle number at time t9, it performs the detection process using images captured at times t6 to t8 and times t10 to t12.

[0048] More specifically, the management device 1 performs the detection process going back in time equivalent to the second cycle P2, starting from time t9, and then performs the detection process sequentially in time equivalent to the second cycle P2 after time t9. Depending on the condition of the vehicle number detected at time t9, the management device 1 may perform the detection process only for either the period before time t9 (times t6 to t8) or the period after time t9 (times t10 to t12), but this will be described in detail later. In the following, the expression "going back by the second cycle P2" may be used, but this is equivalent to going back by the time equivalent to the second cycle P2. In addition, the expression "going forward by the second cycle P2" may be used, but this is equivalent to going forward by the time equivalent to the second cycle P2.

[0049] In other words, if the management device 1 has not acquired the vehicle number, it is unlikely that a vehicle is present, so it performs the detection process in the first period P1, and if it has acquired the vehicle number, it is likely that the vehicle number can be detected even before or after that time, so it performs the detection process in the second period P2.

[0050] As a result, the management device 1 can improve the accuracy of detecting vehicle numbers by performing the detection process at a wide processing cycle (first cycle P1) during periods when there is a low possibility that a vehicle is present, thereby reducing the processing load, and by performing the detection process at a short processing cycle (second cycle P2) during periods when there is a high possibility that a vehicle is present.

[0051] In the above description, the management device 1 has given an example of detecting (obtaining) a vehicle number as a trigger for changing the processing cycle of the detection process from the first cycle P1 to the second cycle P2, but the example of the trigger is not limited to this. The trigger can be any vehicle information related to a vehicle, as long as it indicates a high possibility of the vehicle being present, such as obtaining a WCN or a motion detection function using a camera (a function for detecting a moving vehicle from time-series images).

[0052] Next, a configuration example of the management device 1 according to the embodiment will be described with reference to Fig. 2. Fig. 2 is a functional block diagram showing a configuration example of the management device 1 according to the embodiment. Note that in the block diagram of Fig. 2, only the components necessary for explaining the features of this embodiment are shown as functional blocks, and descriptions of general components are omitted.

[0053] In other words, each component shown in the block diagram of Figure 2 is a functional concept and does not necessarily have to be physically configured as shown. For example, the specific form of distribution and integration of each functional block is not limited to that shown, and all or part of it can be functionally or physically distributed and integrated in any unit depending on various loads, usage conditions, etc.

[0054] 2, the management device 1 includes a control unit 2 and a storage unit 3. The control unit 2 includes an acquisition unit 21, a detection unit 22, an association unit 23, a generation unit 24, and a provision unit 25. The storage unit 3 stores image information 31 and travel history information 32.

[0055] Here, the management device 1 includes, for example, a computer having a central processing unit (CPU), read only memory (ROM), random access memory (RAM), a hard disk drive, input / output ports, and various other circuits.

[0056] The CPU of the computer functions as the acquisition unit 21, detection unit 22, association unit 23, generation unit 24, and provision unit 25 of the control unit 2, for example, by reading and executing a reception program stored in the ROM.

[0057] In addition, at least some or all of the acquisition unit 21, detection unit 22, association unit 23, generation unit 24 and provision unit 25 of the control unit 2 can be configured using hardware such as an ASIC (Application Specific Integrated Circuit) or FPGA (Field Programmable Gate Array).

[0058] The storage unit 3 is configured with a storage device such as a semiconductor element memory, a hard disk drive, etc. The storage unit 3 stores image information 31, travel history information 32, various programs (such as a vehicle detection program), and various information required for processing by the control unit 2.

[0059] Next, the registration information DB 210 stored in the server device 200 will be described with reference to Fig. 3. Fig. 3 is a diagram showing an example of the registration information DB 210. Note that, although the present disclosure shows an example in which the registration information DB 210 is stored in the server device 200, it may also be stored in the management device 1.

[0060] As shown in FIG. 3, the registration information DB 210 includes a "vehicle ID," a "registered vehicle number," and "identification information." The "vehicle ID" is identification information that identifies a vehicle. The "registered vehicle number" is the vehicle number of a vehicle that has been registered in advance. The "identification information" is identification information of the vehicle-mounted device.

[0061] For example, the vehicle identified by the vehicle ID "C1" is associated with the registered vehicle number "Fukuoka, 500, ha, 12-34" and the identification information "WCN#a."

[0062] Next, the travel history information 32 stored in the storage unit 3 of the management device 1 will be described with reference to Fig. 4. Fig. 4 is a diagram showing an example of the travel history information 32. As shown in Fig. 4, the travel history information 32 includes a "travel history ID," "WCN detection information," "vehicle number detection information," and "registration information." Each travel history in the travel history information 32 is generated by the association unit 23, which will be described later.

[0063] "Transit history ID" is identification information that identifies each transit history. "WCN detection information" indicates information related to the detection of identification information, and includes, for example, the items "detection time" and "detected WCN." "Detection time" is information about the time when the identification information was detected (acquired). "Detected WCN" is the detected (acquired) identification information. "Vehicle number detection information" indicates information related to the detection of the vehicle number, and includes, for example, the items "detection time" and "detected vehicle number." "Detection time" is information about the time when the vehicle number was detected (image capture time). "Detected vehicle number" is the detected vehicle number. "Registration information" indicates information registered in the registration information DB210, and includes, for example, the items "registered WCN" and "registered vehicle number." "Registered WCN" is identification information registered in the registration information DB210. "Registered vehicle number" is the vehicle number registered in the registration information DB210.

[0064] Next, returning to FIG. 2, each function of the control unit 2 (acquisition unit 21, detection unit 22, association unit 23, generation unit 24, and provision unit 25) will be described in detail.

[0065] The acquisition unit 21 acquires various types of information. For example, the acquisition unit 21 acquires identification information acquired by the acquisition device 11 from the roadside device 10. Specifically, the acquisition unit 21 acquires identification information acquired by the acquisition device 11 multiple times within the communication range, with the acquisition time associated with the information. The association of the acquisition time with the identification information may be performed by the roadside device 10 or by the acquisition device 11.

[0066] The acquisition unit 21 also acquires images captured by the imaging device 12 from the roadside unit 10. Specifically, the acquisition unit 21 acquires images captured by the imaging device 12 in an imaging period P2, with the images associated with the imaging times.

[0067] The detection unit 22 detects the vehicle license plate number of the vehicle based on the vehicle image acquired by the acquisition unit 21. Note that, when the vehicle number is used as a trigger for changing the processing cycle of the detection process described above, the detection unit 22 also functions as an acquisition unit. That is, the detection unit 22 functioning as an acquisition unit acquires (detects) the vehicle number, which is vehicle information indicating the vehicle, when the vehicle passes through a predetermined area (entrance / exit 110). Specifically, the detection unit 22 detects the license plate number information shown on the license plate as the vehicle number through image analysis. The license plate information is information including items such as a place name, a classification number, hiragana characters, and a serial designated number. Note that the detection unit 22 may detect only a portion of the license plate number information (for example, the serial designated number) as the vehicle number.

[0068] Furthermore, the detection unit 22 functioning as an acquisition unit detects a moving vehicle from the time-series images acquired by the acquisition unit 21, and acquires the detected moving vehicle as vehicle information. For example, the detection unit 22 functioning as an acquisition unit detects a vehicle from an image by template matching, and detects the detected vehicle as a moving vehicle if it is moving in the next image. Furthermore, the detection unit 22 functioning as an acquisition unit may detect a vehicle from an image by template matching, and acquire the detected vehicle as vehicle information.

[0069] Furthermore, when the acquisition unit (acquisition unit 21 or detection unit 22) has not acquired vehicle information, the detection unit 22 performs the detection process of the vehicle number plate at a first period P1 that is longer than the imaging period P2. Furthermore, when the acquisition unit (acquisition unit 21 or detection unit 22) has acquired vehicle information, the detection unit 22 performs the detection process of the vehicle number plate at a second period P2 that is shorter than the first period P1. Note that the second period P2 may be the same as the imaging period P2, or may be a period longer than the imaging period P2 (a period shorter than the first period P1).

[0070] 5 and 6, an example of the detection process by the detection unit 22 will be described. Fig. 5 is a diagram showing an example of the detection process by the detection unit 22. Fig. 6 is a diagram showing another example of the detection process by the detection unit 22.

[0071] Based on the position of the vehicle number detected from the image by the detection process of the first period P1, the detection unit 22 determines whether to perform the detection process of the second period P2 on the image before the detection timing (acquisition timing) of the vehicle number, or to perform the detection process of the second period P2 on the image after the detection timing.

[0072] Specifically, the detection unit 22 first sets a first region R1 and a second region R2 in the image. The first region R1 and the second region R2 are set based on the traveling direction of the vehicle in the image. Specifically, the first region R1 is a region corresponding to the near side of the traveling direction of the vehicle in the image. The second region R2 is a region corresponding to the far side of the traveling direction of the vehicle in the image. Furthermore, the area near the top of the sky region and the area near the bottom of the ground region in the image are not set as the first region R1 and the second region R2. This is because there is a low possibility (no possibility) that a vehicle number plate will be present in the sky region or the ground region. This makes it possible to improve the detection accuracy of the vehicle number plate while reducing the processing load by performing the detection process excluding the sky region and the ground region.

[0073] The positions and sizes of the first region R1 and the second region R2 may be set manually by a user (administrator) or automatically. When set automatically, the detection unit 22 learns the positions of vehicle numbers detected in the past and the traveling direction of the vehicle (the direction of change in the position of the vehicle in the image) and automatically sets the positions and sizes of the first region R1 and the second region R2.

[0074] Next, the detection unit 22 determines whether the vehicle number detected by the detection process in the first period P1 is present in the first region R1 or the second region R2. For example, as shown in the left diagram of FIG. 5, when the position of the vehicle number is present in the first region R1, the detection unit 22 performs the detection process in the second period P2 on images captured after the detection timing of the vehicle number. In other words, since the vehicle number is detected at a position on the near side in the direction of travel, it is highly likely that the vehicle number will be included in images following this image. Therefore, the detection unit 22 performs the detection process in the second period P2 on images captured after the detection timing. Note that the detection unit 22 may perform the detection process in the second period P2 on images captured before the detection timing after completing the detection process on images captured after the detection timing (three images in FIG. 5).

[0075] On the other hand, as shown in the right diagram of Fig. 5, when the position of the vehicle number is in the second region R2, the detection unit 22 performs detection processing in the second cycle P2 on images captured before the detection timing of the vehicle number. In other words, since the vehicle number is detected at a position ahead in the direction of travel, it is highly likely that the vehicle number is included in images preceding this image, and therefore the detection unit 22 performs detection processing in the second cycle P2 on images captured before the detection timing. Note that the detection unit 22 may perform detection processing in the second cycle P2 on images captured after the detection timing after completing detection processing on images captured before the detection timing (three images in Fig. 5).

[0076] In this way, the detection unit 22 can perform detection processing on images in which the vehicle number is likely to be present by determining whether the position of the vehicle number detected in the first period P1 is in the first region R1 or the second region R2, thereby improving the detection accuracy of the vehicle number in the detection processing of the second period P2.

[0077] Next, another example of the detection process will be described with reference to FIG. 6. In FIG. 6, the detection unit 22 sets a first region R1 to a third region R3 in the image. Specifically, the first region R1 to the third region R3 are set based on the traveling direction of the vehicle in the image. Specifically, the first region R1 is a region corresponding to the near side in the traveling direction of the vehicle in the image. The second region R2 is a region corresponding to the far side in the traveling direction of the vehicle in the image. The third region R3 is set at a position sandwiched between the first region R1 and the second region R2. In other words, the third region R3 is a region corresponding to the central position in the traveling direction of the vehicle in the image. Note that, although the first region R1 to the third region R3 have the same region size in FIG. 6, they may have different region sizes.

[0078] Next, the detection unit 22 determines whether the vehicle number detected by the detection process of the first period P1 exists in the first region R1 to the third region R3. If the detection unit 22 determines that the vehicle number exists in the first region R1 or the second region R2, it performs the same detection process as in FIG.

[0079] On the other hand, if the detection unit 22 determines that the vehicle number is present in the third region R3, the detection unit 22 performs the detection process alternately in the second cycle P2 on the images captured before and after the detection timing of the vehicle number. In the example shown in FIG. 6, the detection unit 22 first performs the detection process on the images captured two cycles P2 back from the detection timing of the vehicle number (S1). Next, the detection unit 22 performs the detection process on the images captured two cycles P2 forward from the detection timing of the vehicle number (S2). Next, the detection unit 22 performs the detection process on the images captured two cycles P2 back from the image of S1 (S3). Next, the detection unit 22 performs the detection process on the images captured two cycles P2 forward from the image of S2 (S4). Next, the detection unit 22 performs the detection process on the images captured two cycles P2 back from the image of S3 (S5). Next, the detection unit 22 performs the detection process on the images captured two cycles P2 forward from the image of S4 (S6).

[0080] That is, if the vehicle number plate is located at the center position in the direction of travel at the detection timing of the first period P1, the detection unit 22 performs the detection process alternately on the images before and after the detection timing in the second period P2, since it is highly likely that the vehicle number plate is present in the images before and after this image. In this way, the detection unit 22 can perform the detection process on the images in which the vehicle number plate is highly likely to be present by determining in which of the first region R1 to third region R3 the position of the vehicle number plate detected in the first period P1 is located, thereby improving the detection accuracy of the vehicle number plate in the detection process of the second period P2.

[0081] In addition, in Figure 6, an example is shown in which, in the detection process of the second period P2, the detection process (S1) is performed first on an image that is two periods P2 back from the detection timing, but the detection process may also be performed first on an image that is two periods P2 ahead from the detection timing.

[0082] 6, the detection unit 22 may perform the detection process on both images taken after and before the detection timing only when the vehicle number is present in the third region R3. That is, when the vehicle number is present in the first region R1, the detection unit 22 performs the detection process on only images taken after the detection timing, and when the vehicle number is present in the second region R2, the detection process on only images taken before the detection timing.

[0083] The detection unit 22 determines the final vehicle number detection result based on the detection result of the detection process in the second cycle P2. Specifically, if the vehicle number detected in the second cycle P2 matches the vehicle number detected in the first cycle P1, the detection unit 22 determines the vehicle number as the detection result. Furthermore, the detection unit 22 may determine the vehicle number if a vehicle number matching the vehicle number detected in the first cycle P1 is detected multiple times consecutively in the detection process in the second cycle P2. Furthermore, if there is a vehicle number detected in the second cycle P2 that does not match the vehicle number detected in the first cycle P1, the detection unit 22 may determine the vehicle number that is detected the most times in the first cycle P1 and the second cycle P2 as the final vehicle number. Note that details of the vehicle number determination process will be described later with reference to FIG. 11. The detection unit 22 outputs the determined vehicle number to the association unit 23.

[0084] The associating unit 23 associates the identification information acquired by the acquiring unit 21 with the vehicle number detected by the detecting unit 22. Specifically, the associating unit 23 associates the identification information acquired during a predetermined association period based on the detection time of the vehicle number with the vehicle number.

[0085] The generation unit 24 generates a travel history linked to the vehicle number (detected vehicle number) and identification information (detected WCN) associated by the association unit 23, and stores the travel history in the travel history information 32. Furthermore, if the detected vehicle number and detected WCN are of a registered vehicle, the generation unit 24 links the registered vehicle number and registered WCN to the travel history.

[0086] The providing unit 25 provides various information to a user (administrator) via a display unit of a terminal device that is the management device 1. For example, the providing unit 25 displays, on the display unit, an image captured by the imaging device 12 and a screen including information related to the detection process.

[0087] Here, an example of a screen provided by the providing unit 25 will be described with reference to Fig. 7. Fig. 7 is a diagram showing an example of a screen provided by the providing unit 25. Fig. 7 shows a screen W displayed on a display unit of a terminal device that is the management device 1. Note that the screen W provided by the providing unit 25 is not limited to being displayed on the display unit of the management device 1, and may also be displayed on another terminal device that is connected to the management device 1 via wireless or wired communication.

[0088] 7 shows an example of a screen W including information relating to the entrance side of the entrance / exit 110. As shown in Fig. 7, the screen W displays information relating to the gate status, buttons relating to gate opening / closing control, and information relating to vehicles that have passed through the entrance (the number of detected vehicles and the passage history at the bottom of the screen). Also, as shown in Fig. 7, the screen W displays a video screen M that displays images captured by the imaging device 12 as live images in real time.

[0089] Here, an example of a screen displayed on the video screen M will be described with reference to Fig. 8. Fig. 8 is a diagram showing an example of a screen displayed on the video screen M. In Fig. 8, "normal analysis in progress" indicates a state in which detection processing is being performed in the first cycle P1, and "intensive analysis in progress" indicates a state in which detection processing is being performed in the second cycle P2. In Fig. 8, images M1 to M5 are displayed in this order on the video screen M.

[0090] Image M1 shown in Fig. 8 indicates an image displayed during normal analysis. That is, image M1 indicates an image for detection processing performed by detection unit 22 in first period P1. Then, when a vehicle number plate is detected in second region R2 (see Fig. 5) in image M1, providing unit 25 displays image M2, which is taken two periods earlier than the detection timing, which is the time of image M1. Note that when a vehicle number plate is not detected in image M1 (if no vehicle is present in image M1), providing unit 25 displays an image captured in the next imaging period (or the next first period P1) following image M1.

[0091] 8 is an image being analyzed intensively, and is an image captured at a time going back (rewinding) the second cycle P2 from image M1. That is, image M2 represents an image for detection processing performed in the second cycle P2 by detection unit 22. When detection processing by detection unit 22 is completed for image M2, provision unit 25 displays image M3, which was captured at a time going back the second cycle P2 from image M2.

[0092] Then, when the detection unit 22 completes the detection process for image M3, the providing unit 25 receives a notification from the detection unit 22 indicating whether the detection process has been performed sufficiently retroactively. Specifically, the detection unit 22 determines whether an image exists going back a second period P2 from image M3, and notifies the providing unit 25 of the determination result. More specifically, the detection unit 22 determines whether the image going back a second period P2 from image M3 is the same as the image going back a first period P1 from image M1. In other words, the detection unit 22 determines whether the image going back a second period P2 from image M3 is an image for which vehicle number detection has been completed through the detection process for the first period P1. If the image going back a second period P2 from image M3 is an image for which vehicle number detection has been completed through the detection process for the first period P1, the detection unit 22 notifies the providing unit 25 that the detection process has been performed sufficiently retroactively. When the providing unit 25 receives a notification indicating that the detection process has been performed sufficiently retroactively, it displays image M4 shown in FIG. 8.

[0093] Next, image M4 shown in Fig. 8 is an image undergoing intensive analysis, and is an image captured at a time two cycles P2 ahead (delayed) from image M1. That is, image M4 represents an image for detection processing performed by detection unit 22 in second cycle P2. Then, when detection processing by detection unit 22 is completed for image M4, providing unit 25 displays image M5 undergoing normal analysis. Note that image M5 is the latest image captured by imaging device 12.

[0094] In this way, the providing unit 25 displays the images M2 to M4 for which the detecting unit 22 performs the detection process (intensive analysis) in the second cycle P2, so that the user can check the images that are being intensively analyzed.

[0095] Next, the procedure of the process executed by the management device 1 according to the embodiment will be described with reference to Figs. 9 to 11. Fig. 9 is a flowchart showing the procedure of the overall process executed by the management device 1 according to the embodiment. Fig. 10 is a flowchart showing the procedure of the vehicle number detection process executed by the management device 1 according to the embodiment. Fig. 11 is a flowchart showing the procedure of the vehicle number determination process executed by the management device 1 according to the embodiment.

[0096] As shown in FIG. 9, first, the control unit 2 of the management device 1 acquires the identification information acquired by the acquisition device 11 (step S101).

[0097] Next, the control unit 2 performs a process of detecting the vehicle number based on the image captured by the imaging device 12 (step S102). The detailed procedure of the detection process will be described with reference to FIG.

[0098] Next, the control unit 2 performs an association process to associate the acquired identification information with the detected vehicle number (step S103).

[0099] Next, the control unit 2 generates a travel history linked with the associated identification information and vehicle number, and stores it in the storage unit 3 as travel history information 32, thereby registering the travel history (step S104).

[0100] Next, the control unit 2 controls the opening and closing of the gate to allow the vehicle to pass (step S105), and ends the process.

[0101] 10 is the process executed in step S102 described above, and is executed for each second cycle P2. Specifically, the control unit 2 first acquires an image captured by the imaging device 12 (step S201), and stores the image as image information 31 in the storage unit 3 (step S202).

[0102] Next, the control unit 2 determines whether it is time for the detection process of the first cycle P1 (step S203). If it is time for the first cycle P1 (step S203: Yes), the control unit 2 performs the detection process of the vehicle number (step S204). If it is not time for the first cycle P1 (step S203: No), the control unit 2 returns to step S201.

[0103] Next, the control unit 2 determines whether or not a vehicle number has been detected from the image (step S205). If the control unit 2 detects the vehicle number (step S205: Yes), it performs a vehicle number determination process described later in FIG. 11, and determines whether or not the vehicle number has been determined as a result of the vehicle number determination process (step S206). If the control unit 2 determines the vehicle number as a result of the vehicle number determination process (step S206: Yes), it ends the process. Note that if the control unit 2 has not detected the vehicle number (step S205: No), it returns to step S201.

[0104] On the other hand, if the vehicle number cannot be determined (is undetermined) as a result of the vehicle number determination process (step S206: No), the control unit 2 determines whether the detected vehicle number is present in the second area R2 (step S207).

[0105] Next, if the vehicle number is present in the second region R2 (step S207: Yes), the control unit 2 performs a detection process using images going back a second period P2 (step S208).

[0106] Next, when the control unit 2 detects the vehicle number by the detection process of S208, it performs a vehicle number determination process described later in Fig. 11, and determines whether the vehicle number has been determined as a result of the vehicle number determination process (step S209). When the vehicle number has been determined as a result of the vehicle number determination process (step S209: Yes), the control unit 2 ends the process.

[0107] On the other hand, if the vehicle number cannot be determined (is undetermined) as a result of the vehicle number determination process (step S209: No), the control unit 2 determines whether it has gone back to the period just before the first period P1 (whether it has gone back sufficiently) (step S210).

[0108] If the control unit 2 has traced back to the first cycle P1 (step S210: Yes), it performs detection processing using an image from the second cycle P2 ahead from the detection timing (step S211). If the control unit 2 has not traced back to the first cycle P1 (step S210: No), it returns to step S208.

[0109] Next, when the control unit 2 detects the vehicle number by the detection process of S211, it performs a vehicle number determination process described later in Fig. 11, and determines whether the vehicle number has been determined as a result of the vehicle number determination process (step S212). When the control unit 2 determines that the vehicle number has been determined as a result of the vehicle number determination process (step S212: Yes), it ends the process.

[0110] On the other hand, if the vehicle number cannot be determined (is undetermined) as a result of the vehicle number determination process (step S212: No), the control unit 2 determines whether or not the process has progressed to just before the first cycle P1 (step S213), and if the process has not progressed to just before the first cycle P1 (step S213: No), the control unit 2 returns to step S211.

[0111] On the other hand, if the process has progressed to just before the first cycle P1 (step S213: Yes), the control unit 2 determines that the vehicle number is undetermined (step S214) and ends the process.

[0112] In step S207, if the vehicle number is present in the first region R1 (step S207: No), the control unit 2 performs a detection process using an image of the second period P2 ahead (step S215).

[0113] Next, when the control unit 2 detects the vehicle number by the detection process of S215, it performs a vehicle number determination process described later in Fig. 11, and determines whether the vehicle number has been determined as a result of the vehicle number determination process (step S216). When the vehicle number has been determined as a result of the vehicle number determination process (step S216: Yes), the control unit 2 ends the process.

[0114] On the other hand, if the vehicle number cannot be determined (is undetermined) as a result of the vehicle number determination process (step S216: No), the control unit 2 determines whether or not the vehicle number has progressed to just before the first cycle P1 (step S217).

[0115] If the control unit 2 has proceeded to just before the first cycle P1 (step S217: Yes), the control unit 2 performs detection processing using an image taken two cycles P2 back from the detection timing (step S218). If the control unit 2 has not proceeded to just before the first cycle P1 (step S217: No), the control unit 2 returns to step S215.

[0116] Next, when the control unit 2 detects the vehicle number by the detection process of S218, it performs a vehicle number determination process described later in Fig. 11, and determines whether the vehicle number has been determined as a result of the vehicle number determination process (step S219). When the control unit 2 determines that the vehicle number has been determined as a result of the vehicle number determination process (step S219: Yes), it ends the process.

[0117] On the other hand, if the vehicle number cannot be determined (is undetermined) as a result of the vehicle number determination process (step S219: No), the control unit 2 determines whether it has gone back to the period just before the first period P1 (step S220), and if it has not gone back to the period just before the first period P1 (step S220: No), it returns to step S218.

[0118] On the other hand, if the control unit 2 has traced back to the period immediately before the first period P1 (step S220: Yes), it determines that the vehicle number is undetermined (step S214) and ends the process.

[0119] Next, the determination process shown in Fig. 11 will be described. The determination process shown in Fig. 11 is the process executed in the above-mentioned steps S206, S209, S212, S216, and S219. Specifically, it is assumed that the control unit 2 detects the vehicle number by the detection process (step S301).

[0120] Next, the control unit 2 determines whether the reliability of the detected vehicle number is equal to or greater than a threshold (step S302). The reliability is, for example, information indicating the degree of detection accuracy of the vehicle number, such as the rate of match between the numbers and characters of the vehicle number and a template.

[0121] If the reliability is equal to or greater than the threshold (step S302: Yes), the control unit 2 determines whether the same vehicle number exists in the most recent vehicle number confirmation list (step S303). Here, the most recent vehicle number confirmation list will be described with reference to FIG.

[0122] FIG. 12 is a diagram showing an example of the most recent vehicle number confirmation list. The most recent vehicle number confirmation list is stored, for example, in the memory unit 3 of the management device 1. The most recent vehicle number confirmation list is a list of vehicle numbers that have been confirmed by confirmation processing most recently (for example, within about 5 minutes). In other words, the most recent vehicle number confirmation list is a list of vehicle numbers that have been confirmed by confirmation processing performed when a vehicle passes through the entrance / exit 110. Note that in the most recent vehicle number confirmation list, once the time that a vehicle would have passed through the entrance / exit 110 has passed, the vehicle number information detected from that vehicle is deleted.

[0123] As shown in FIG. 12, the most recent confirmed vehicle number list has items such as "confirmed ID," "confirmed vehicle number," "analysis time (first)," and "analysis time (last)."

[0124] "Confirmed ID" is identification information that identifies the confirmed vehicle number. "Confirmed vehicle number" is information about the confirmed vehicle number. "Analysis time (first)" is the time when the confirmed vehicle number was first confirmed (the time of the first confirmation process). "Analysis time (last)" is the time when the confirmed vehicle number was last confirmed (the time of the last confirmation process).

[0125] The most recent vehicle number confirmed list shown in Figure 12 deletes the information for the corresponding "confirmed ID" from the most recent vehicle number confirmed list when the time that a vehicle would have passed through entrance / exit 110 has passed, that is, when the time that has elapsed since the "analysis time (last)" has reached the aforementioned time (approximately 5 minutes).

[0126] Returning to FIG. 11, if the same vehicle number as the detected vehicle number exists in the most recent vehicle number confirmation list (step S303: Yes), the control unit 2 updates the "analysis time (last)" of the most recent vehicle number confirmation list to the latest time (the latest time when the confirmation process was performed (step S304)), confirms it with the vehicle number detected in step S301 (step S305), and ends the confirmation process.

[0127] On the other hand, if the reliability is less than the threshold value (step S302: No), the control unit 2 determines that the vehicle number is undetermined (step S306) and ends the process.

[0128] Furthermore, if the same vehicle number does not exist in the most recent vehicle number confirmation list (step S303: No), the control unit 2 determines whether a similar vehicle number has been detected within a short period of time (step S307). Note that "within a short period of time" means within the time that the vehicle number is listed in the most recent vehicle number confirmation list. In other words, in step S307, the control unit 2 determines whether a similar vehicle number is included in the most recent vehicle number confirmation list. Note that a similar vehicle number is, for example, a vehicle number in which only one number or letter on the license plate differs, in other words, a vehicle number whose vehicle number matching rate is equal to or exceeds a threshold value (for example, 90%).

[0129] If a similar vehicle number is detected within a short period of time (step S307: Yes), the control unit 2 determines that the vehicle number is erroneously detected, and treats the vehicle number as undetermined in step S306.

[0130] On the other hand, if no similar vehicle number is detected within a short period of time (step S307: No), the control unit 2 determines that the vehicle number has a low probability of being falsely detected, treats it as a new vehicle number, and determines whether the new vehicle number matches the registered vehicle number (step S308).

[0131] If the new vehicle number does not match the registered vehicle number (step S308: No), the control unit 2 adds the new vehicle number to the vehicle number detection list (step S309), and determines whether the same vehicle number has been detected multiple times within a short period of time by referring to the vehicle number detection list (step S310). Here, the vehicle number detection list will be described with reference to FIG.

[0132] 13 is a diagram showing an example of the vehicle number detection list. The vehicle number detection list is stored, for example, in the storage unit 3 of the management device 1. The vehicle number detection list is a list to which an entry is added each time the vehicle number detected in S301 differs from the registered vehicle number in S308.

[0133] As shown in FIG. 13, the vehicle number detection list has items such as "detection ID," "detected vehicle number," and "detection time." "Detection ID" is identification information that identifies the detected vehicle number. "Detected vehicle number" is information about the detected vehicle number. "Detection time" is the time when the detected vehicle number was detected.

[0134] Returning to Figure 11, if the same vehicle number is detected multiple times within a short period of time (step S310: Yes), the control unit 2 adds the detected vehicle number to the most recent vehicle number confirmation list by issuing a new confirmed ID (step S311), and confirms the vehicle number in step S305.

[0135] On the other hand, if the new vehicle number matches the registered vehicle number (step S308: Yes), the control unit 2 updates the most recent vehicle number confirmation list in step S304. Also, if the same vehicle number is not detected multiple times within a short period of time (step S310: No), the control unit 2 treats the vehicle number as unconfirmed in step S306.

[0136] As described above, according to one embodiment of the present disclosure, the vehicle detection device (management device 1) includes the detection unit 22 and an acquisition unit (acquisition unit 21 and detection unit 22). The detection unit 22 performs a detection process to detect the vehicle license plate number of a vehicle passing through a predetermined area (entrance / exit 110) at a predetermined cycle from images continuously captured by the camera (imaging device 12) in accordance with an imaging cycle P2. The acquisition unit acquires vehicle information indicating the vehicle when the vehicle passes through the predetermined area. If the acquisition unit has not acquired vehicle information, the detection unit 22 performs the detection process at a first cycle P1 longer than the imaging cycle P2. If the acquisition unit has acquired vehicle information, the detection unit 22 performs the detection process at a second cycle P2 shorter than the first cycle P1. This enables the vehicle detection device to reduce the processing load while maintaining detection accuracy.

[0137] In the above-described embodiment, the detection process for the second period P2 is performed on all images during the first period P1 (images from times t6 to t8 and times t10 to t12 shown in FIG. 1B), but this is not limiting. In other words, it is possible to arbitrarily set how far back or how far forward the detection process for the second period P2 should be performed on among the images during the first period P1.

[0138] Furthermore, the detection unit 22 may use machine learning to determine how far back or how far forward the images to be processed for the detection process of the second cycle P2. Specifically, the detection unit 22 receives environmental information indicating the environment of the entrance / exit 110 (environmental information indicating the environment of a predetermined area) as input and generates in advance a model that outputs a target period (a period to go back, a period to postpone) for the detection process of the second cycle P2. The detection unit 22 inputs the environmental information into the generated model, thereby determining the target period for the detection process of the second cycle P2.

[0139] Here, the environmental information includes, for example, the traveling speed of the vehicle when passing through the entrance / exit 110, the detection accuracy of the vehicle number in each area in the image, weather information, time information, and the like.

[0140] For example, the slower the traveling speed, the more likely it is that the vehicle number plate will be detected in the first region R1 (the closer side in the traveling direction), so the model parameters are adjusted so that the forward period is longer than the retroactive period as the target period for the detection process. Note that the traveling speed can be detected based on time-series images captured by the imaging device 12. Alternatively, the traveling speed may be acquired from an on-board device of the vehicle.

[0141] Furthermore, the detection accuracy of vehicle number plates in each region of the image is measured by dividing the image into multiple regions (for example, dividing the first region R1 and the second region R2 into three regions in the direction of travel, for a total of six regions), and tallying the number of detections for each region. The more detections there are in a region, the more likely it is that the vehicle number plate can be detected, and the model parameters are adjusted to output a target period that includes regions with a high number of detections.

[0142] In addition, weather information and time information are used to adjust the model parameters so that, for example, if there is an area in the image where the license plate is reflected by sunlight on a sunny day and the vehicle number cannot be detected, the model will determine a target period that avoids that area.

[0143] In this way, the detection unit 22 determines the target period for the detection process of the second cycle P2 using a model that has learned environmental information, thereby being able to determine a target period with high vehicle license plate detection accuracy. In other words, the detection unit 22 can improve the vehicle license plate detection accuracy.

[0144] Furthermore, among the processes described in the above embodiments, all or part of the processes described as being performed automatically can be performed manually, or all or part of the processes described as being performed manually can be performed automatically using a known method. In addition, the information including the processing procedures, specific names, various data, and parameters shown in the above documents and drawings can be changed as desired unless otherwise specified. For example, the various information shown in each drawing is not limited to the information shown in the drawings.

[0145] Furthermore, the components of each device shown in the figure are conceptual functional components and do not necessarily have to be physically configured as shown in the figure. In other words, the specific form of distribution and integration of each device is not limited to that shown in the figure, and all or part of them can be functionally or physically distributed and integrated in any unit depending on various loads, usage conditions, etc.

[0146] Furthermore, the present invention is not limited to the above-described embodiments, and various modifications are possible within the scope of the claims. For example, the technical scope of the present invention also includes configurations obtained by appropriately combining the above-described embodiments in an area where the processing content is not contradictory. Furthermore, the order of each step shown in the flowcharts and sequence diagrams of the above-described embodiments can be changed as appropriate. [Explanation of symbols]

[0147] 1 Management device 2. Control Unit 3 Storage section 10 Roadside unit 11 Acquisition device 12 Imaging device 21 Acquisition Department 22 Detection unit 23 Association section 24 Generation part 25 Providing Department 31 Image information 32 Traffic history information 110 Entrance / exit 200 Server device 210 Registration Information DB S Entrance / Exit Control System

Claims

1. a detection unit that performs a detection process to detect, at a predetermined cycle, the vehicle license plate number of a vehicle passing through a predetermined area from images captured continuously in accordance with the imaging cycle of the camera; an acquisition unit that acquires vehicle information indicating the vehicle when the vehicle passes through the predetermined area; Equipped with The detection unit When the acquisition unit has not acquired the vehicle information, the detection process is performed at a first period longer than the imaging period, and when the acquisition unit has acquired the vehicle information, the detection process is performed at a second period shorter than the first period. Vehicle detection device.

2. The acquisition unit Acquire a vehicle number as the vehicle information, The detection unit The detection process is performed every second period at least either before or after the timing of acquiring the vehicle number. The vehicle detection device of claim 1 .

3. The acquisition unit a first area corresponding to the near side in the traveling direction of the vehicle and a second area corresponding to the far side in the traveling direction of the vehicle are set in the image, and it is determined in which area the vehicle number is located; The detection unit If the vehicle number is present in the first area, the detection process is performed for the images after the acquisition timing in each second period, and if the vehicle number is present in the second area, the detection process is performed for the images before the acquisition timing in each second period. The vehicle detection device according to claim 2 .

4. The detection unit If the vehicle number is present in the first area, the detection process is performed for the images after the acquisition timing every second period, and then the detection process is performed for the images before the acquisition timing every second period; If the vehicle number is present in the second area, the detection process is performed for the image before the acquisition timing at every second period, and then the detection process is performed for the image after the acquisition timing at every second period. The vehicle detection device according to claim 3 .

5. The acquisition unit setting a third area between the first area and the second area, and determining in which area the vehicle number is located; The detection unit If the vehicle number is present in the third area, the detection process is performed alternately for each of the images after the acquisition timing and the images before the acquisition timing in each second period. The vehicle detection device according to claim 3 .

6. A storage unit is further provided for storing a most recent vehicle number confirmation list, which is a list of vehicle numbers detected by the detection unit during a period when the vehicle passes through the predetermined area and confirmed as detection results, The detection unit If the vehicle number acquired by the acquisition unit matches the vehicle number included in the most recent vehicle number confirmation list, the detection process is not performed for each second period, and the vehicle number acquired by the acquisition unit is confirmed as a detection result. The vehicle detection device according to claim 2 .

7. The detection unit If the vehicle number acquired by the acquisition unit matches a pre-registered vehicle number, the detection process is not performed for each second period, and the vehicle number acquired by the acquisition unit is determined as a detection result. The vehicle detection device according to claim 2 .

8. The second period is the same period as the imaging period. The vehicle detection device of claim 1 .

9. The detection unit A target period of the detection process to be performed in each second cycle is determined based on environmental information indicating an environment of the predetermined area. The vehicle detection device of claim 1 .

10. 1. A computer-implemented vehicle detection method, comprising: a detection step of detecting the vehicle license plate number of a vehicle passing through a predetermined area at a predetermined period from images captured continuously in accordance with the imaging period of the camera; an acquisition step of acquiring vehicle information indicating the vehicle when the vehicle passes through the predetermined area; Including, The detecting step When the acquisition step does not acquire the vehicle information, a detection process is performed at a first period longer than the imaging period, and when the acquisition step acquires the vehicle information, the detection process is performed at a second period shorter than the first period. Vehicle detection methods.

11. A vehicle detection program executed by a computer, a detection step for detecting the vehicle license plate number of a vehicle passing through a predetermined area at a predetermined period from images captured continuously in accordance with the imaging period of the camera; an acquisition step of acquiring vehicle information indicating the vehicle when the vehicle passes through the predetermined area; on the computer, The detection procedure comprises: When the acquisition procedure has not acquired the vehicle information, a detection process is performed in a first cycle that is longer than the imaging cycle, and when the acquisition procedure has acquired the vehicle information, the detection process is performed in a second cycle that is shorter than the first cycle. Vehicle detection program.

Images