Unauthorized passage determination device and unauthorized passage determination method

Abstract

To provide an unauthorized passage determination device that reduces a time and human resources required for unauthorized passage determination and improves efficiency of the unauthorized passage determination.SOLUTION: An unauthorized passage determination device according to an embodiment includes a data input unit, an unauthorized passage determination unit, and an output unit. The data input unit inputs video data output from a camera that captures a vehicle passage lane. The unauthorized passage determination unit determines unauthorized passage of a determination target vehicle based on characteristic data related to unauthorized passage vehicles obtained from the video data input during a first period, and the characteristic data related to the determination target vehicle obtained from video data input during a second period after the first period. The output unit outputs warning information based on a determination result of the unauthorized passage.SELECTED DRAWING: Figure 5

Description

[Technical Field]

[0001] An embodiment of the present invention relates to an unauthorized passage determination device and an unauthorized passage determination method. [Background technology]

[0002] Electronic Toll Collection Systems (ETC) are widely used on toll roads across the country. The ETC system uses wireless communication at each toll gate to enable non-stop toll collection without requiring vehicles to stop at entrance and exit toll gates. At ETC entrance toll gates, an antenna installed corresponding to the lane of the entrance toll gate communicates wirelessly with the on-board units of vehicles traveling in the lane of the entrance toll gate. At ETC exit toll gates, an antenna installed corresponding to the lane of the exit toll gate also communicates wirelessly with the on-board units of vehicles traveling in the lane of the exit toll gate via the lane of the entrance toll gate. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2017-208050 [Patent Document 2] Japanese Patent Application Laid-Open No. 201-1475 [Patent Document 3] Japanese Patent Application Publication No. 2022-174470 Summary of the Invention [Problem to be solved by the invention]

[0004] On expressways where ETC systems have been introduced, there are frequent cases of fraudulent traffic at toll booths where drivers try to avoid paying tolls. There are several types of traffic that can be defined as fraudulent traffic, such as "forcing your way through a lane without paying the toll" or "falsely declaring the entrance that serves as the starting point for toll calculations when exiting from an exit toll booth."

[0005] Currently, each person in charge at the operation company visually reviews large amounts of video data to determine whether a vehicle is passing illegally. Each person also identifies illegal vehicles based on the vehicle type and license plate number that they visually read.

[0006] Therefore, it takes a lot of time and human resources to detect illegal passage, and problems that depend on the individual (such as overlooking illegal passage and variations in the accuracy of illegal passage detection) may occur. [Means for solving the problem]

[0007] The problem to be solved by the present invention is to provide an illegal passage determination device and an illegal passage determination method that reduce the time and human resources required for illegal passage determination and make the illegal passage determination more efficient.

[0008] An illegal passage determination device according to an embodiment includes a data input unit, an illegal passage determination unit, and an output unit. The data input unit inputs video data output from a camera that captures the vehicle's traffic lane. The illegal passage determination unit determines the illegal passage of the vehicle to be determined based on feature data related to the illegally passing vehicle obtained from the video data input during a first period and feature data related to the vehicle to be determined obtained from video data input during a second period after the first period. The output unit outputs warning information based on the result of the illegal passage determination. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is a diagram illustrating an example of a schematic configuration of an ETC according to an embodiment. [Figure 2] FIG. 2 is a block diagram showing an example of a schematic configuration of the road entrance device according to the embodiment. [Figure 3] FIG. 3 is a block diagram showing an example of a schematic configuration of the exit roadside device according to the embodiment. [Figure 4] FIG. 4 is a diagram illustrating the concept of fraudulent passage determination by the upper server according to the embodiment. [Figure 5]FIG. 5 is a diagram illustrating an example of functional blocks of an upper server according to the embodiment. [Figure 6] FIG. 6 is a diagram illustrating an example of the fraudulent passage determination process of the upper server according to the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, embodiments will be described with reference to the drawings.

[0011] [composition] FIG. 1 is a diagram illustrating an example of a schematic configuration of an ETC according to an embodiment. We will explain an ETC system, which is an example of a toll fee presentation system, as an example. The ETC system comprises a centrally located host server 1, an information output device 6 that outputs any illegal passage detected by the host server 1, an entrance toll gate that handles entrance processing to the toll road, and an exit toll gate that handles exit processing from the toll road.

[0012] A data processing device 21 installed at the entrance toll gate and a data processing device 22 installed at the exit toll gate are connected via dedicated lines to the upper server 1. The upper server 1 is an example of an illegal passage determination device.

[0013] At the entrance toll gate, a data processing device 21, a toll gate server 31, entrance lane servers 41 corresponding to the number of lanes, and an entrance roadside device 51 on the roadside of each lane are installed.

[0014] At the exit toll gate, a data processing device 22, a toll gate server 32, and exit lane servers 42 corresponding to the number of lanes are installed, and exit roadside devices 52 are installed on the roadside of the lanes.

[0015] First, a description will be given of the host server 1. The host server 1 is made up of one or more computers, and includes a control unit 11, a storage unit 12, a communication unit 13, and a user interface .

[0016] The control unit 11 has a processor such as a CPU, a memory, an interface, etc. The CPU is a processor that realizes various processing functions by executing programs, and is connected to each component that constitutes the upper server 1. The memory includes a non-volatile memory that stores the programs executed by the CPU, and a volatile memory that temporarily stores data, etc. The control unit 11 calculates the toll based on the toll table, entrance identification information, and exit identification information, and generates toll information indicating the toll. The toll may also be calculated at the exit toll gate.

[0017] The storage unit 12 includes a storage medium that can read and write data at high speed, such as a main memory, and also includes a large-capacity storage device such as an HDD or SSD. The storage unit 12 stores programs executed by the CPU of the control unit 11 and a fee table.

[0018] The communication unit 13 is a communication interface that allows the upper server 1 to communicate with the outside via a network. The user interface 14 includes a display device and an input device, and receives input from the toll road operator (person in charge). The input information is stored in the memory unit 12.

[0019] Next, each device installed at the toll booth will be explained.

[0020] FIG. 2 is a block diagram showing an example of a schematic configuration of the road entrance device according to the embodiment. The data processing device 21 is configured with one or more computers, and includes a control unit 211, a storage unit 212, and a communication unit 213. The control unit 211, the storage unit 212, and the communication unit 213 are basic components of a computer.

[0021] Next, the toll booth server 31 is configured with one or more computers, and includes a control unit 311, a storage unit 312, and a communication unit 313. The control unit 311, the storage unit 312, and the communication unit 313 are basic components of a computer.

[0022] Next, the entrance lane server 41 manages the entrance roadside device 51 and controls the entrance roadside device 51 based on signals from the entrance roadside device 51. The entrance lane server 41 is composed of one or more computers, and has a control unit 411, a memory unit 412, and a communication unit 413.

[0023] The control unit 411 controls the operation of the start controller of the entrance roadside unit 51 based on the vehicle detection signal from the entrance roadside unit 51. The control unit 411 has a processor such as a CPU, a memory, an interface, etc. The CPU is a processor that realizes various processing functions by executing programs. The memory includes a non-volatile memory that stores the programs executed by the CPU, and a volatile memory that temporarily stores data, etc.

[0024] The storage unit 412 stores entrance identification information (for example, a toll gate number) that identifies the entrance toll gate. The storage unit 412 includes a storage medium that can read and write data at high speed, such as a main memory, and also includes a large-capacity storage device. The storage unit 412 stores programs that the CPU of the control unit 411 executes.

[0025] The communication unit 413 is a communication interface that communicates with the outside via a network. As the vehicle travels from the upstream lane to the downstream lane, the communication unit 413 transmits information received from the vehicle's on-board device via a first antenna 5131 on the upstream side of the lane of the entrance roadside device 51 to the upper server 1, and also transmits information from the upper server 1 to the vehicle's on-board device via a second antenna 5132 on the downstream side of the lane.

[0026] Next, we will explain the entrance roadside device 51. The entrance roadside device 51 includes a communication unit 511, a control unit 512, a vehicle communication unit 513, a vehicle detection unit 514, an axle detector 515, a start controller 516, a roadside display 517, and a camera 518.

[0027] In addition, when dealing with a free flow in which the vehicle is not decelerated, the start controller 516 may be excluded from the configuration of the entrance roadside device 51.

[0028] The communication unit 511 is a communication interface for communicating with the entrance lane server 41 .

[0029] The control unit 512 is a control unit that controls the entire entrance roadside unit 51 and performs data processing, etc. The control unit 512 has a CPU, memory, an interface, etc. The CPU is a processor that realizes various processing functions by executing programs. The memory includes various types of memory, such as non-volatile memory that stores programs executed by the CPU and volatile memory that temporarily stores data.

[0030] The vehicle communication unit 513 includes a first antenna 5131 and a second antenna 5132. The first antenna 5131 communicates with an on-board device of a vehicle traveling in an area upstream of the lane, and receives on-board device information, user information, etc. The second antenna 5132 communicates with an on-board device of a vehicle traveling in an area downstream of the lane, and transmits entrance passage information to the on-board device.

[0031] Vehicle detection unit 514 is an optical sensor or the like, and is provided along the lane of the entrance toll gate to detect vehicles passing through the lane of the entrance toll gate. For example, vehicle detection unit 514 is provided corresponding to second antenna 5132, and detects vehicles passing near second antenna 5132.

[0032] Axle detector 515 is a pressure sensor such as a tread sensor embedded in the lane of the entrance toll gate, and detects the axles of vehicles passing through the lane of the entrance toll gate. For example, axle detector 515 is provided corresponding to first antenna 5131, and detects the number of axles of vehicles passing near first antenna 5131.

[0033] For example, the control unit 512 detects the period during which one vehicle passes based on the vehicle detection signal output from the vehicle detection unit 514, and detects the number of axles of one vehicle based on the axle detection signal output from the axle detector 515 during the period during which one vehicle passes. The control unit 512 determines the type of vehicle based on the detected number of axles.

[0034] The departure controller 516 is a control unit that controls the entry of vehicles into the toll road by controlling the opening and closing of a departure control bar provided in the lane at the entrance toll gate. Based on the vehicle detection signal output from the vehicle detection unit 514 and the results of communication with the vehicle via the first antenna 5131, the departure controller 516 closes the departure control bar to prevent the vehicle from passing when the vehicle is to be stopped temporarily or when the vehicle is to be prohibited from passing (entering). In addition, when the departure controller 516 permits the vehicle to pass (enter), it opens the departure control bar to allow the vehicle to pass.

[0035] In addition, when dealing with a free flow in which the vehicle is not decelerated, the start controller 516 may be excluded from the configuration of the entrance roadside device 51.

[0036] The roadside display 517 displays information such as guidance to passengers in vehicles traveling in the entrance lane. The roadside display 517 is installed facing vehicles traveling in the entrance lane. For example, the roadside display 517 displays toll information provided at the entrance that indicates the toll to the exit.

[0037] The camera 518 captures the lane of travel of the vehicle and outputs video data. The communication unit 511 transmits the video data, and the host server 1 receives the video data directly or indirectly.

[0038] Next, the configuration of the exit toll gate will be described.

[0039] FIG. 3 is a block diagram showing an example of a schematic configuration of the exit roadside device according to the embodiment. The data processing device 22 and the toll gate server 32 are similar to the data processing device 21 and the toll gate server 31 at the entrance toll gate, and therefore a description thereof will be omitted.

[0040] The exit lane server 42 manages the exit roadside device 52 and controls the exit roadside device 52 based on signals from the exit roadside device 52. The exit lane server 42 is made up of one or more computers, and includes a control unit 421, a storage unit 422, and a communication unit 423. The hardware configuration of the exit lane server 42 is similar to that of the entrance lane server 41, and a description of the common parts will be omitted.

[0041] When calculating a toll at an exit toll gate, the memory unit 422 stores a toll table transmitted from the upper server 1. The control unit 421 calculates the toll based on the toll table and information transmitted from the vehicle's on-board device, and outputs the toll. The information transmitted from the vehicle's on-board device includes on-board device information, vehicle information, contract information, and entrance passage information. The vehicle information includes the vehicle type, license plate information, and vehicle power type.

[0042] Next, a description will be given of the exit roadside device 52. The exit roadside device 52 includes a communication unit 521, a control unit 522, a vehicle communication unit 523, a vehicle detection unit 524, a departure controller 526, a roadside display 527, and a camera 528.

[0043] The vehicle communication unit 523 includes an antenna 5231. The hardware configuration of the exit roadside device 52 is similar to that of the entrance roadside device 51, and a description of the common parts will be omitted.

[0044] The antenna 5231 of the exit roadside unit 52 will be described.

[0045] The exit roadside device 52 communicates with the vehicle's onboard device in one communication area using an antenna 5231. Note that the exit roadside device 52 may also communicate with the vehicle's onboard device in two communication areas using first antennas 5131 and 5132, similar to the entrance roadside device 51.

[0046] The antenna 5231 communicates with the on-board device of the vehicle passing through the exit toll gate, receives on-board device information, vehicle information, contract information, and entrance passage information, and transmits the toll calculated by the exit lane server 42. The roadside display 527 displays the toll calculated by the exit lane server 42.

[0047] The camera 528 captures the lane of travel of the vehicle and outputs video data. The communication unit 521 transmits the video data, and the host server 1 receives the video data directly or indirectly.

[0048] FIG. 4 is a diagram illustrating the concept of fraudulent passage determination by the upper server according to the embodiment. The camera 518 of the entrance roadside device 51 photographs the traffic lane near the entrance gate of the entrance toll gate and outputs video data. The camera 528 of the exit roadside device 52 photographs the traffic lane near the exit gate of the exit toll gate and outputs video data.

[0049] For the sake of explanation, video data captured during a learning period (first period) corresponding to the setting of the learning mode will be referred to as past video data. Also, video data captured during a fraudulent vehicle detection period (second period) corresponding to the setting of the fraudulent vehicle detection mode will be referred to as new video data. For example, during the learning period, past video data is transmitted to the upper server 1. The upper server 1 receives past video data one after another via the communication unit 13 and stores it in the learning database of the memory unit 12. During a fraudulent vehicle detection period after the learning period, new video data is transmitted to the upper server 1. The upper server 1 receives the new video data via the communication unit 13.

[0050] The host server 1 displays an image based on the past video data stored in the learning database via the user interface 14. While visually checking the displayed image of the past video data, the operator inputs, via the user interface 14, the type of vehicle, vehicle number, vehicle manufacturer, vehicle model, and illegal passage information of the illegally passing vehicle, which are at least part of the characteristic data regarding the illegally passing vehicle.

[0051] Here, the characteristic data related to illegally passing vehicles will be explained. The characteristic data related to illegally passing vehicles includes the traffic pattern of the illegally passing vehicle, attribute data of the illegally passing vehicle, and illegal passing information. The traffic pattern is the deceleration and acceleration pattern of the illegally passing vehicle. The attribute data of the illegally passing vehicle includes the vehicle type, vehicle number, vehicle manufacturer, and vehicle model of the illegally passing vehicle. The illegal passing information includes whether or not illegal passing occurred, the date and time of the illegal passing, the number of times the illegal passing occurred, etc.

[0052] The upper server 1 uses the control unit 11 to add part of the feature data related to illegally passing vehicles to the past video data to generate correct answer data, and registers the correct answer data in the correct answer database in the memory unit 12. In addition, the upper server 1 uses the control unit 11 to learn illegal passage patterns from the feature data related to illegally passing vehicles, and registers the learning results in the illegal passage database in the memory unit 12.

[0053] The host server 1 acquires characteristic data on the vehicle to be judged from the new video data using the control unit 11, judges whether the vehicle to be judged is passing illegally based on the characteristic data on the illegally passing vehicle and the characteristic data on the vehicle to be judged, and outputs warning information based on the judgment result of illegal passage. Furthermore, the host server 1 can judge unknown illegal passage using the learning results registered in the illegal passage database using the control unit 11. The host server 1 outputs detailed data including the judgment result of illegal passage to the information output device 6. Furthermore, the host server 1 outputs warning information based on the judgment result of illegal passage to the information output device 6 using the communication unit 13.

[0054] The detailed data is automatically created when fraudulent passage occurs, and includes the gate number, the name of the entrance toll gate, the time of passing through the entrance toll gate, information such as the vehicle type, the vehicle number, and whether or not there has been a history of fraudulent passage in the past.

[0055] FIG. 5 is a diagram illustrating an example of functional blocks of an upper server according to the embodiment. 1, the upper server 1 includes a control unit 11, a memory unit 12, a communication unit 13, and a user interface 14. The control unit 11 includes a data input unit 1111, a data storage unit 1112, a number assignment unit 1114, a fraudulent passage information assignment unit 1115, a fraudulent passage pattern registration unit 1117, a number detection unit 1118, an accuracy calculation unit 1119, a fraudulent passage determination unit 1120, a warning unit 1121, and a detail output unit 1122. The memory unit 12 includes a learning database (DB) 1213, a fraudulent passage database (DB) 1216, a detail database (DB) 1223, etc.

[0056] The data input unit 1111 inputs video data output from the camera 518 or 528 that captures the vehicle traffic lane. The video data includes passing vehicles. The data input unit 1111 inputs the video data as past video data for learning in response to the setting of the learning mode. Alternatively, the data input unit 1111 inputs the video data as new video data for detecting unauthorized vehicles in response to the setting of the unauthorized vehicle detection mode.

[0057] The data storage unit 1112 stores the past video data received from the data input unit 1111 in the learning database 1213. The learning database 1213 stores the past video data.

[0058] The number assignment unit 1114 retrieves past video data from the learning database 1213, assigns attribute data such as the vehicle type, vehicle number, vehicle manufacturer, and vehicle model input via the user interface 14 to the past video data, and registers the past video data with the assigned attribute data in the learning database 1213. For example, the number assignment unit 1114 assigns the attribute data in association with a predetermined playback time of the past video data.

[0059] The fraudulent passage information assignment unit 1115 retrieves past video data to which attribute data has been assigned from the learning database 1213, assigns fraudulent passage information including the fraudulent passage judgment results input via the user interface 14 to the past video data, and registers the past video data to which the fraudulent passage information, etc. has been assigned in the fraudulent passage database 1216.

[0060] The fraudulent passage pattern registration unit 1117 analyzes and learns fraudulent passage patterns based on past video data to which fraudulent passage information has been added, and registers the learned results of the fraudulent passage patterns in the fraudulent passage database 1216. Possible fraudulent passage patterns include a combination of a certain deceleration pattern and a certain acceleration pattern. Also, possible fraudulent passage patterns include a combination of a certain deceleration pattern, a certain acceleration pattern, and the driver's behavior. Furthermore, possible fraudulent passage patterns include the relationship between the contents of the electronic bulletin board and the driver's behavior.

[0061] The license plate number detection unit 1118 detects the vehicle type, manufacturer, model, and license plate number of the vehicle contained in the new video data received from the data input unit 1111 based on the past video data registered in the learning database 1213 .

[0062] The accuracy calculation unit 1119 calculates the accuracy of the vehicle type, vehicle license plate number, vehicle manufacturer, and vehicle model detected by the license plate detection unit 1118. The accuracy calculation unit 1119 performs accuracy detection on the video, using the most frequent value of the accuracy of each frame included in the video data as a reference. In other words, the accuracy calculation unit 1119 detects the vehicle type, vehicle license plate number, vehicle manufacturer, and vehicle model for the image of each frame, and uses the most frequent values ​​of the vehicle type, vehicle license plate number, vehicle manufacturer, and vehicle model as the detection results. Images with accuracy lower than a certain value may be excluded from the determination process.

[0063] The illegal passage determination unit 1120 determines the illegal passage of the vehicle to be determined based on the feature data of the illegally passing vehicle obtained from the past video data and the feature data of the vehicle to be determined obtained from the new video data. That is, the illegal passage determination unit 1120 determines the illegal passage of the vehicle to be determined based on the similarity between the passage pattern of the illegally passing vehicle and the passage pattern of the vehicle to be determined. For example, the illegal passage determination unit 1120 determines the illegal passage of the vehicle to be determined based on the similarity between the deceleration and acceleration patterns of the illegally passing vehicle and the deceleration and acceleration patterns of the vehicle to be determined. The illegal passage determination unit 1120 determines that the vehicle to be determined is illegally passing if the similarity exceeds a reference value, and determines that the vehicle to be determined is passing normally if the similarity does not exceed the reference value.

[0064] Furthermore, the fraudulent passage determination unit 1120 may determine the fraudulent passage of the vehicle to be determined based on the output timing of display information corresponding to the passage timing of the vehicle to be determined and the deceleration and acceleration pattern of the vehicle to be determined. The passage timing of the vehicle to be determined is detected based on a signal from the vehicle detection unit 514 or 524, and the roadside display 517 or 527 outputs display information corresponding to the passage timing. The fraudulent passage determination unit 1120 may determine that the passage of the vehicle to be determined is fraudulent when the similarity between the deceleration and acceleration pattern of the fraudulent passing vehicle and the deceleration and acceleration pattern of the vehicle to be determined exceeds a reference value and further when the deceleration and acceleration pattern of the vehicle to be determined is detected based on the output timing of the display information.

[0065] The attention generating unit 1121 transmits information based on the determination result of the illegal passage determination unit 1120 to the information output device 6. For example, the attention generating unit 1121 transmits warning information based on the determination result of illegal passage to the information output device 6. The attention generating unit 61 of the information output device 6 outputs the warning information to alert the user.

[0066] The detail output unit 1122 outputs the detail data and associates the new video data including the illegally passing vehicle with the detail data. Furthermore, the detail output unit 1122 registers data associating the new video data including the illegally passing vehicle with the detail data in the detail database 1223.

[0067] FIG. 6 is a diagram illustrating an example of the fraudulent passage determination process of the upper server according to the embodiment. As shown in S11, the data input unit 1111 receives video data output from the camera 518 or 528 that captures the traffic lane of the vehicle.

[0068] First, the learning process corresponding to the setting of the learning mode will be described.

[0069] As indicated by YES in S12, the data input unit 1111 inputs, as past video data, video data captured during a learning period corresponding to the setting of the learning mode.

[0070] As shown in S13, the data storage unit 1112 stores the past video data received from the data input unit 1111 in the learning database 1213.

[0071] The number assignment unit 1114 retrieves past video data from the learning database 1213, and the operator inputs attribute data including the vehicle type, vehicle number, vehicle manufacturer, and vehicle model via the user interface 14 while visually checking the display image of the retrieved past video data.

[0072] As shown in S15, the number assigning unit 1114 assigns the attribute data input via the user interface 14 to the past video data, and registers the past video data to which the attribute data has been assigned in the learning database 1213.

[0073] As indicated by YES in S16, the operator visually checks the display image of the extracted past video data to determine whether or not the traffic is illegal. If the traffic is illegal, the operator inputs illegal traffic information including the illegal traffic history via the user interface 14.

[0074] As shown in S17, the fraudulent passage information assigning unit 1115 assigns the fraudulent passage information, including the fraudulent passage history, to the past video data input via the user interface 14. In other words, the past video data to which the attribute data and feature data, including the fraudulent passage information, have been assigned is registered in the learning database 1213.

[0075] As shown in S18, the fraudulent passage pattern registration unit 1117 analyzes the fraudulent passage patterns of the fraudulent vehicles based on past video data to which feature data including attribute data and fraudulent passage information has been added, learns the fraudulent passage patterns, and registers the fraudulent passage patterns in the fraudulent passage database 1216.

[0076] As described above, the fraudulent passage information assigning unit 1115 and the fraudulent passage pattern registering unit 1117 register feature data related to fraudulent passing vehicles obtained from past video data in the learning database 1213 and the fraudulent passage database 1216. The feature data includes attribute data of the fraudulent passing vehicles, fraudulent passage information, and fraudulent passage patterns.

[0077] Next, the unauthorized vehicle detection process corresponding to the setting of the vehicle detection mode will be described.

[0078] As indicated by NO in S12, the data input unit 1111 inputs, as new video data, video data captured during a fraudulent vehicle detection period corresponding to the setting of the fraudulent vehicle detection mode.

[0079] As shown in S20, the number detection unit 1118 detects attribute data including the vehicle type, vehicle number, vehicle manufacturer, vehicle model, etc. of the vehicle to be judged contained in the new video data received from the data input unit 1111 based on the learning data registered in the learning database 1213.

[0080] The accuracy calculation unit 1119 calculates the accuracy of the attribute data including the vehicle type, vehicle number, vehicle manufacturer, vehicle model, etc. detected by the number detection unit 1118. The accuracy calculation unit 1119 calculates the accuracy at regular time intervals within the video and calculates the average.

[0081] As shown in S22, the fraudulent passage determination unit 1120 determines whether the vehicle to be determined corresponds to a fraudulent passage vehicle registered in the fraudulent passage database based on the attribute data of the fraudulent passage vehicle registered in the fraudulent passage database and the attribute data of the detected vehicle to be determined, such as the vehicle type, vehicle number, vehicle manufacturer, and vehicle model, and further determines whether the identified vehicle to be determined has a fraudulent passage history. For example, the fraudulent passage determination unit 1120 determines that the vehicle to be determined corresponds to a fraudulent passage vehicle registered in the fraudulent passage database when the vehicle number included in the attribute data of the fraudulent passage vehicle matches a vehicle number included in the attribute data of the vehicle to be determined that has a certainty equal to or greater than a reference value.

[0082] As shown in S23, if the target vehicle has a history of illegal passage, the warning unit 1121 transmits information about the illegal passage to the information output device 6.

[0083] As shown in S24, the fraudulent passage determination unit 1120 determines whether the passage pattern of the fraudulent passing vehicle matches the passage pattern of the vehicle to be determined. If the similarity between the passage pattern of the fraudulent passing vehicle and the passage pattern of the vehicle to be determined exceeds a reference value, the fraudulent passage determination unit 1120 determines that the passage pattern of the fraudulent passing vehicle matches the passage pattern of the vehicle to be determined, and if the similarity does not exceed the reference value, it determines that they do not match.

[0084] For example, the fraudulent passage determination unit 1120 reads the vehicle number of the fraudulent passage and the pattern of the fraudulent passage from the fraudulent passage database, determines whether the read number and pattern of the fraudulent passage match the number and pattern contained in the new video data provided by the accuracy calculation unit 1119, and if they match, determines that the passage is fraudulent, and if they do not match, determines that the passage is normal.

[0085] As shown in S25, in response to the determination of fraudulent passage, the warning unit 1121 transmits information regarding the fraudulent passage to the information output device 6.

[0086] As shown in S26, the detail output unit 1122 outputs the detail data, associates the new video data including the illegally passing vehicle with the detail data, and registers the data associating the new video data including the illegally passing vehicle with the detail data in the detail database 1223.

[0087] As described above, the host server according to the embodiment provides the following effects. (1) The upper server detects unauthorized passage from new video data, which reduces the number of operators required. (2) The host server learns information about vehicle types and license plates from past video data and uses that learning data to detect vehicle types and license plates for new video data, thereby improving detection accuracy. (3) The upper server learns illegal passage, derives illegal passage patterns, and determines illegal passage based on the illegal passage patterns, making it possible to determine illegal passage even for vehicles that have not engaged in illegal passage in the past. (4) The host server learns vehicle make and model in addition to vehicle type and license plate number. This makes it possible to detect vehicle make and model, which not only increases the accuracy of detecting illegal passage but also comes in handy in the future when it becomes necessary to change toll fees based on classifications beyond vehicle type.

[0088] Although several embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be embodied in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their modifications are included within the scope and spirit of the invention, and are also included in the scope of the invention and its equivalents as defined in the claims. [Explanation of symbols]

[0089] 1...Upstream server 6...Information output device 11...Control unit 12...Storage section 13…Communications Department 14...User Interface 21, 22...Data processing device 31, 32...Toll booth server 41...Entry lane server 42...Exit lane server 51…Entrance roadside equipment 52...Exit roadside device 1111...Data input section 1112...Data storage section 1114...Number Assignment Department 1115...Illegal Traffic Information Department 1117...Illegal Traffic Pattern Registration Department 1118...Number plate detection unit 1119...Accuracy calculation section 1120...Illegal Passing Detection Department 1121…Warning section 1122...Detail output section 1213...Learning database 1216...Fraudulent Passing Database 1223...Detailed database

Claims

1. a data input unit that inputs video data output from a camera that captures images of vehicle traffic lanes; an illegal passage determination unit that determines illegal passage of a target vehicle based on feature data on the illegally passing vehicle obtained from video data input during a first period and feature data on the target vehicle obtained from video data input during a second period after the first period; an output unit that outputs warning information based on the result of the illegal passage determination; An illegal passage determination device comprising:

2. The characteristic data regarding the illegally passing vehicle includes a traffic pattern of the illegally passing vehicle, the characteristic data relating to the target vehicle includes a traffic pattern of the target vehicle; the fraudulent passage determination unit determines the fraudulent passage of the target vehicle based on a similarity between a passage pattern of the fraudulent passing vehicle and a passage pattern of the target vehicle; The fraudulent passage determination device according to claim 1.

3. the fraudulent passage determination unit determines the fraudulent passage of the fraudulent passing vehicle based on the similarity between the deceleration and acceleration pattern of the fraudulent passing vehicle and the deceleration and acceleration pattern of the target vehicle; The unauthorized passage determination device according to claim 2.

4. the fraudulent passage determination unit determines the fraudulent passage of the vehicle to be determined based on an output timing of display information corresponding to a passage timing of the vehicle to be determined and a deceleration and acceleration pattern of the vehicle to be determined. The unauthorized passage determination device according to claim 3.

5. the characteristic data regarding the illegally passing vehicle includes attribute data of the illegally passing vehicle; the feature data relating to the target vehicle includes attribute data of the target vehicle; the fraudulent passage determination unit identifies the vehicle to be determined based on attribute data of the fraudulent passage vehicle and attribute data of the vehicle to be determined; The fraudulent passage determination device according to claim 1.

6. the fraudulent passage determination unit identifies the vehicle to be determined when a vehicle number included in the attribute data of the fraudulent passage vehicle matches a vehicle number included in the attribute data of the determination target vehicle with a certainty equal to or greater than a reference value; The fraudulent passage determination device according to claim 5.

7. a fraudulent passage information assigning unit that assigns fraudulent passage information to predetermined video data in response to input of fraudulent passage information for the predetermined video data input during a first period; a fraudulent passage pattern registration unit that registers the passage pattern of the fraudulent passing vehicle based on the predetermined video data in a fraudulent passage database, The unauthorized passage determination device according to claim 2.

8. A method for determining fraudulent passage executed by a fraudulent passage determination device, comprising: Input video data output from a camera capturing the vehicle's traffic lane, determining whether the subject vehicle is passing illegally based on feature data relating to the illegally passing vehicle obtained from the video data input during a first period and feature data relating to the subject vehicle obtained from the video data input during a second period after the first period; A method for determining illegal passage, which outputs warning information based on the result of the illegal passage determination.

Images