Detection device, detection system, and computer program

Abstract

This detection device comprises: a detection unit for acquiring sensor information from a sensor that senses an object by transmitting an electromagnetic wave toward a road and receiving the electromagnetic wave reflected by the object, and detecting, on the basis of the sensor information, an event set in advance; a selection unit for selecting, according to the details of the event detected by the detection unit, a camera among a plurality of cameras installed on the road, said camera capturing an image related to the event; and an instruction unit for instructing the camera selected by the selection unit to capture an image.

Description

Technical Field

[0001] The present disclosure relates to a detection device, a detection system, and Computer program . This application claims priority based on Japanese Application No. 2021-116600 filed on July 14, 2021, and incorporates all the descriptions set forth in the above-mentioned Japanese application.

Background Art

[0002] Conventionally, a system is known that installs a camera on a road where vehicles pass and monitors the road conditions based on images captured by the camera.

[0003] Patent Document 1 describes a system that detects a vehicle that ignores a signal at an intersection and photographs the detected vehicle with a camera. This system includes an intersection panoramic camera that captures a panorama of the intersection, a vehicle photographing camera that photographs a specific vehicle that has entered the intersection, and a speed sensor that detects a vehicle entering the intersection at a speed above a set value. When the speed sensor detects a vehicle (signal-ignoring candidate vehicle) entering the intersection at a speed above the set value during a red signal, the system processes the video of the vehicle photographing camera to detect the signal-ignoring candidate vehicle. When a signal-ignoring candidate vehicle is detected, the system converts the video of the vehicle photographing camera into a plurality of frames of still images and records them. As a result, the vehicle number (license plate) and the driver of the vehicle that ignored the signal at the intersection are recorded as still images.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

[0005] The detection device disclosed herein comprises: a detection unit that detects an object by transmitting electromagnetic waves to a road and receiving the electromagnetic waves reflected by the object, and acquires sensor information from the sensor and detects a preset event based on the acquired sensor information; a selection unit that selects a camera from among a plurality of cameras installed on the road to take an image related to the event according to the content of the event detected by the detection unit; and an instruction unit that instructs the camera selected by the selection unit to take an image.

[0006] The detection method disclosed herein is a detection method comprising the steps of: obtaining sensor information from a sensor that detects an object by transmitting electromagnetic waves to a road and receiving the electromagnetic waves reflected by the object; detecting a preset event based on the acquired sensor information; selecting a camera from among a plurality of cameras installed on the road to take an image related to the event, according to the content of the detected event; and instructing the selected camera to take an image. [Brief explanation of the drawing]

[0007] [Figure 1] Figure 1 is a schematic diagram showing an example of the installation of the detection system according to the embodiment. [Figure 2] Figure 2 is a schematic perspective view showing the sensor unit according to this embodiment. [Figure 3] Figure 3 is a block diagram showing the functional configuration of the detection system according to the embodiment. [Figure 4] Figure 4 is a flowchart showing the control structure of the program executed by the detection device according to the embodiment. [Figure 5] Figure 5 is a flowchart showing the control structure of the program executed by the detection device according to the embodiment. [Figure 6] Figure 6 is a flowchart showing the control structure of the program executed by the camera according to the embodiment. [Figure 7] Figure 7 is a sequence diagram showing an example of a detection method performed by the detection system according to the embodiment. [Figure 8] Figure 8 is a block diagram illustrating the processing performed by the trained discrimination model in the modified example. [Figure 9] Figure 9 is a block diagram illustrating the process of generating training data related to the modified example. [Figure 10] Figure 10 is a flowchart showing the sequence of operations performed by the detection device related to the modification. [Modes for carrying out the invention]

[0008] [Problems the invention aims to solve] In the system described in Patent Document 1, if a vehicle suspected of running a red light is not detected when the video from the vehicle camera is processed, the video from the vehicle camera is not converted into a still image. In other words, even if an event such as running a red light is detected, if the vehicle camera does not properly capture the vehicle suspected of running a red light, there is a problem in that information about the vehicle suspected of running a red light (vehicle number, etc.) is not recorded.

[0009] In view of these challenges, this disclosure provides a detection device, a detection system, and a detection device that can more accurately record image information relating to detected events. Computer program The purpose is to provide.

[0010] [Effects of the invention] According to this disclosure, image information related to detected events can be recorded more accurately.

[0011] [Description of Embodiments in this Disclosure] Embodiments of this disclosure include, in essence, at least the following:

[0012] (1) The detection device of the present disclosure is a detection device comprising: a detection unit that detects an object by transmitting electromagnetic waves to a road and receiving the electromagnetic waves reflected by the object, and which acquires sensor information from a sensor that detects an object based on the acquired sensor information, a selection unit that selects a camera from among a plurality of cameras installed on the road to take an image related to the event according to the content of the event detected by the detection unit, and an instruction unit that instructs the camera selected by the selection unit to take an image.

[0013] According to the detection device of this disclosure, depending on the content of the detected event, a camera is selected from among multiple cameras installed on the road to capture images related to the detected event. Because a camera suitable for capturing images related to the detected event can be selected, image information related to the detected event can be recorded more accurately.

[0014] (2) There may be multiple pre-set events, and the detection unit may detect one or more events from among the multiple pre-set events based on the sensor information. This allows the camera to be selected to capture images related to the detected event, so that image information related to the detected event can be recorded appropriately.

[0015] (3) The preset plurality of events may include events that may occur in the target area where the sensor acquires sensor information. This allows for the recording of suitable image information for events that may occur in the target area where sensor information is acquired.

[0016] (4) The pre-set events may include at least one of the following: exceeding the legal speed limit or designated speed, speeding by a vehicle on the road, driving the wrong way on the road, parking a vehicle on the road, road congestion, and the presence of fallen objects on the road. Such events are of high importance to record. Therefore, with this configuration, image information relating to events of high importance to record can be suitably recorded.

[0017] (5) When the detection unit detects road driving with a speed exceeding the limit by a vehicle as the event, the selection unit may select, from the plurality of cameras, a camera that captures an area downstream of the target area from which the sensor acquires the sensor information with respect to the driving direction of the road as the imaging target area.

[0018] By configuring in this way, it is possible to more reliably capture a vehicle that is driving on the road with a speed exceeding the limit.

[0019] (6) When the detection unit detects reverse driving of a vehicle on the road as the event, the selection unit may select, from the plurality of cameras, a camera that captures an area upstream of the target area from which the sensor acquires the sensor information with respect to the driving direction of the road as the imaging target area.

[0020] By configuring in this way, it is possible to more reliably capture a vehicle that is driving in reverse.

[0021] (7) The instruction unit may determine, as the imaging condition of the camera selected by the selection unit, either a first imaging condition for imaging at a predetermined number of frames or a second imaging condition for imaging at a number of frames greater than the predetermined number of frames according to the event detected by the detection unit, and may give an instruction to image according to the determined imaging condition.

[0022] By configuring in this way, it is possible to image with a more suitable number of frames according to the event, so that the detailed information of the event can be detected more accurately based on the image.

[0023] (8) If the detection unit detects, as a preset event, a vehicle parked on the road, traffic congestion on the road, or the presence of an object on the road, the instruction unit may determine the first shooting conditions as the shooting conditions for the camera selected by the selection unit, or may instruct the camera to shoot according to the determined first shooting conditions. Also, if the detection unit detects, as a preset event, a vehicle speeding on the road exceeding the legal speed limit or designated speed limit, or a vehicle driving the wrong way on the road, the instruction unit may determine the second shooting conditions as the shooting conditions for the camera selected by the selection unit, or may instruct the camera to shoot according to the determined second shooting conditions.

[0024] For events involving moving vehicles, such as speeding or driving the wrong way, shooting with a higher frame rate ensures that the moving vehicle is more reliably included in the image. Conversely, for events involving stationary or relatively slow-moving objects, such as parked cars, traffic jams, or fallen objects, shooting with a lower frame rate can save data capacity.

[0025] (9) The system may further include a detailed detection unit that detects detailed information of the event detected by the detection unit based on the image captured by the camera selected by the selection unit.

[0026] (10) If the detection unit detects, as a preset event, a vehicle speeding on the road exceeding the legal speed limit or designated speed limit, a vehicle driving the wrong way on the road, or a vehicle parked on the road, the detailed detection unit may detect, as detailed information, information relating to the license plate of the vehicle in question.

[0027] (11) The detection system of the present disclosure is a detection system comprising the sensor, a plurality of the cameras, and any of the detection devices described in (1) to (10).

[0028] (12) The detection method of the present disclosure is a detection method comprising the steps of: obtaining sensor information from a sensor that detects an object by transmitting electromagnetic waves to a road and receiving the electromagnetic waves reflected by the object; detecting a preset event based on the acquired sensor information; selecting a camera from among a plurality of cameras installed on the road to take an image related to the event according to the content of the detected event; and instructing the selected camera to take an image.

[0029] According to the detection method disclosed herein, a camera is selected according to the content of the detected event, thereby enabling more accurate recording of image information related to the detected event.

[0030] [Details of the embodiments of this disclosure] The embodiments of this disclosure will be described in detail below with reference to the drawings.

[0031] On roads, various incidents can occur, including illegal parking, objects falling from vehicles, speeding, driving against traffic, and traffic congestion. These incidents are likely to lead to serious accidents. Therefore, it is desirable to record image information related to such incidents, for example, to confirm the circumstances at the time the incident occurred.

[0032] The detection system according to this embodiment acquires sensor information from sensors installed on the road and detects the occurrence of these events by processing the acquired sensor information. The detection system further acquires (records) detailed information about the event by instructing a camera to take a picture based on the detection result.

[0033] The location and subject matter of a camera's image will vary depending on the nature of the event (type of event, location, etc.). For example, if a sensor detects a fallen object on a road, the camera needs to photograph the object and determine what it is based on the image. In this case, the camera should photograph the location where the sensor detected the object, and it is preferable for the camera to zoom in on that location to capture details of the object.

[0034] Furthermore, if a vehicle traveling the wrong way on the road is detected by the sensor, it is preferable that the vehicle be photographed by a camera and information regarding the vehicle's license plate be detected based on the image. In this case, the locations that the camera should photograph are the location where the sensor detected the vehicle and a location located upstream in the direction of traffic (i.e., the location that the vehicle traveling the wrong way will pass after the sensor detects it). For this reason, it is preferable to operate another camera located upstream in the direction of traffic in addition to the camera that photographs the detected location.

[0035] Therefore, the detection system according to this embodiment selects a camera from among multiple cameras installed on the road to capture images related to the detected event, depending on the content of the event. As a result, even when various events occur on the road, and the location of each event and the matters to be recorded differ, the detection system according to this embodiment accurately records the situation of each event using a camera based on the detection results of each event.

[0036] <Overall configuration of the detection system> Figure 1 is a schematic diagram showing an example of the installation of the detection system 10 according to this embodiment. The detection system 10 includes a plurality of detection devices 20a, 20b and a plurality of sensor units 30a, 30b, 30c. It is preferable that the detection devices 20a, 20b each have the same configuration. Unless otherwise specified, the detection devices 20a, 20b will simply be referred to as "detection device 20". It is preferable that the sensor units 30a, 30b, 30c each have the same configuration. Unless otherwise specified, the sensor units 30a, 30b, 30c will simply be referred to as "sensor unit 30". In Figure 1, two detection devices 20 and three sensor units 30 are shown as examples, but the number of detection devices 20 and sensor units 30 included in the detection system 10 is not particularly limited.

[0037] The detection device 20 is a device that detects events based on sensor information from the sensor unit 30. The detection device 20 functions as an integrated processing device that processes sensor information from the sensor unit 30, controls the sensor unit 30, etc., and transmits information with other detection devices. The detection device 20 is connected to the sensor unit 30 via wired or wireless communication. In this embodiment, for example, the detection device 20a controls sensor units 30a and 30b, and the detection device 20b controls, for example, sensor unit 30c. The detection devices 20a and 20b are connected via a telecommunications network N1.

[0038] The detection device 20 and the sensor units 30 may correspond in a one-to-many relationship, as in detection device 20a, or in a one-to-one relationship, as in detection device 20b. Furthermore, one detection device 20 may control all the sensor units 30 included in the detection system 10.

[0039] The detection device 20 and the sensor unit 30 are installed on or near the roadway, in a location facing the roadway (collectively referred to as "Road R1"). Road R1 is, for example, an expressway (national expressway). Road R1 is not particularly limited as long as it is a road on which vehicles travel, and may be a national highway, a prefectural road, or any other road. In addition to the area on which vehicles can normally travel, Road R1 may also include areas on the shoulder and emergency parking areas where vehicles can enter in an emergency, as well as a median strip.

[0040] In Figure 1, arrow AR1 indicates the direction of vehicle traffic on road R1. Road R1 is, for example, one-way, and vehicle traffic is permitted only in the direction AR1. In the following explanation, the area downstream of direction AR1 will be simply referred to as "downstream" as appropriate, and the area upstream of direction AR1 will be simply referred to as "upstream" as appropriate.

[0041] Road R1 is provided with support posts 6a, 6b, ... at predetermined intervals (for example, every 100m to 300m). The detection device 20a is installed at the bottom of support post 6a, and the sensor units 30a and 30b are installed at the top of support post 6a. The detection device 20b is installed at the bottom of support post 6b, and the sensor unit 30c is installed at the top of support post 6b.

[0042] Sensor unit 30 is a unit for detecting events on road R1. Sensor unit 30a detects events in the first region A1, sensor unit 30b detects events in the second region A2, and sensor unit 30c detects events in the third region A3. The first to third regions A1 to A3 are regions included in road R1. The regions set for each sensor unit 30 do not have to overlap with other regions, as in the first region A1, or they may overlap with other regions, as in the second region A2 and third region A3. In this embodiment, the target regions for event detection are arranged in the order of first region A1, second region A2, and third region A3 from upstream.

[0043] The detection device 20 communicates with the management device 200 via the telecommunications network N1. The management device 200 is a device that manages multiple detection devices 20. This management device 200 is located, for example, in the traffic control center TC1.

[0044] <Sensor unit configuration> Figure 2 is a schematic perspective view of the sensor unit 30a. The sensor unit 30a includes a housing 31a, a sensor 40a, and a camera 50a. In this embodiment, the sensor 40a and the camera 50a are housed in a single housing 31a. However, the sensor 40a and the camera 50a may be housed in separate housings.

[0045] Sensor units 30b and 30c have the same configuration as sensor unit 30a. Specifically, sensor unit 30b has a housing (not shown) and a sensor 40b and a camera 50b housed in the housing. Sensor unit 30c also has a housing (not shown) and a sensor 40c and a camera 50c housed in the housing. The housings of sensor units 30a, 30b, and 30c, sensors 40a to 40c, and cameras 50a to 50b are preferably the same configuration, and if no particular distinction is made, they are simply referred to as "housing 31," "sensor 40," and "camera 50."

[0046] Sensor 40 includes a millimeter-wave radar for measuring the position, direction, and speed of an object by radiating electromagnetic waves in the millimeter-wave band (20-300 GHz) toward the object and receiving and processing the reflected waves. For example, FMCW (Frequency Modulated Continuous Wave) is used as the modulation method for the millimeter-wave radar. Sensor 40 has a transmitting unit that emits electromagnetic waves toward the road R1, a receiving unit that receives electromagnetic waves (reflected waves) reflected by the road R1 (or an object on the road R1), and a processing circuit.

[0047] The processing circuit detects the distance to an object whose reflected wave intensity is above a predetermined threshold, the direction of that object, and the velocity of that object. Specifically, the processing circuit calculates the distance from the sensor 40 to the object by measuring the time from the transmission of the electromagnetic wave to the reception of the reflected wave. The receiving unit includes multiple receiving antennas, and the processing circuit calculates the direction of the object relative to the sensor 40 based on the phase difference of the reflected wave resulting from the time difference when the multiple receiving antennas receive the reflected wave. The processing circuit further calculates the velocity of the object relative to the sensor 40 based on the Doppler shift of the received electromagnetic wave.

[0048] The sensor 40 transmits the data of the object's position (distance and direction) and velocity obtained in this manner to the detection device 20 as sensor information D1. The sensor 40 may also include other object detection sensors such as a LiDAR.

[0049] Sensor 40 may be a camera (imaging sensor) that photographs the road R1 using visible light or infrared light. In this case, camera 50 may have both the function of sensor 40 for detecting the presence and type of event, and the function of camera 50 for detecting detailed information about the event. Also, sensor 40 may be a different camera from camera 50.

[0050] Camera 50 is an imaging device for recording detailed information of events detected by sensor 40. For example, under normal circumstances, camera 50 captures a panoramic view of the target area, and when an event is detected, it records detailed information of that event. Camera 50 includes a movable part 51 that can change the shooting direction, a zoom lens 52 that can change the focal length, and an image sensor 53 that converts optical information into electronic signals. Camera 50 may acquire images (still images) one by one in response to commands from detection device 20, or it may acquire multiple images as a video at a predetermined number of frames in response to commands from detection device 20. Furthermore, camera 50 may have a light-emitting unit that emits visible light or infrared light (e.g., strobe flash).

[0051] In this embodiment, the area captured by the camera 50 includes the area where the sensor 40 detects an event. For example, if the sensor 40a detects an event in the first area A1, the camera 50a captures the area including the first area A1. A camera 50 that captures the area including the area detected by the sensor 40 in this way is referred to as "the camera 50 corresponding to the sensor 40". In this embodiment, the camera 50 corresponding to the sensor 40a is "camera 50a", and the camera 50 corresponding to the sensor 40b is "camera 50b".

[0052] <Configuration of the detection device> Figure 3 is a block diagram showing the functional configuration of the detection system 10. Figure 3 shows the functional configuration of the detection device 20a in detail, while the functional configuration of the detection device 20b is the same as that of the detection device 20a and is therefore not shown.

[0053] The detection device 20 (20a) detects events occurring on road R1 based on sensor information D1 transmitted from sensor 40. The detection device 20 is substantially a computer and has a control unit 21, a storage unit 22, and a communication interface that functions as a communication unit 23. The control unit 21 includes an arithmetic unit (processor). The arithmetic unit includes, for example, a CPU (Central Processing Unit). The arithmetic unit may further include a GPU (Graphics Processing Unit). The storage unit 22 includes a main memory unit and an auxiliary memory unit. The main memory unit includes, for example, RAM (Random Access Memory). The auxiliary memory unit includes, for example, an HDD (Hard Disk Drive) or an SSD (Solid State Drive). The detection device 20 realizes the functions of each of the units 24 to 27 described later by the control unit 21 (arithmetic unit) executing a computer program stored in the storage unit 22.

[0054] The control unit 21 has a detection unit 24, a selection unit 25, an instruction unit 26, and a detailed detection unit 27 as functional units. Each of these functional units 24 to 27 may be implemented by the same processing area in the control unit 21, or by separate processing areas. For example, one CPU may implement the functions of both the detection unit 24 and the detailed detection unit 27, or a separate CPU may be provided to implement the functions of the detection unit 24 and the detailed detection unit 27.

[0055] The detection unit 24 detects a predetermined event on road R1 based on sensor information D1 acquired from sensor 40. The storage unit 22 stores a selection table for each of several types of events, which associates the content of the event with the camera 50 to be used for shooting and the shooting conditions. The selection unit 25 refers to the selection table and selects a camera 50 from among several cameras 50 to be used to shoot an image Im1 related to the event, according to the content of the event detected by the detection unit 24. The instruction unit 26 instructs the camera 50 selected by the selection unit 25 to start shooting. The detailed detection unit 27 detects detailed event information D3 based on the image Im1 captured by the camera 50.

[0056] The storage unit 22 stores computer programs, sensor information D1, image Im1, detailed information D3, selection tables, and other parameters. The communication unit 23 transmits and receives various information to and from other detection devices 20 and management devices 200 via the telecommunications network N1.

[0057] <Detection of events by the detection unit 24> The detection unit 24 is configured to detect multiple types of events based on sensor information from the sensor 40. The multiple types of events to be detected include speeding, driving in the wrong direction, parking (illegal parking), falling objects, and traffic congestion by the vehicle V1.

[0058] The detection unit 24 has the function of performing predetermined preprocessing on sensor information D1 from the sensor 40, and the function of executing event detection processing to detect events based on the data obtained by the preprocessing. The preprocessing includes clustering processing and tracking processing, etc.

[0059] Clustering is a process that recognizes an object (e.g., a vehicle V1) by combining multiple reflected wave points contained in sensor information D1 into a single combined entity. This process makes it possible to recognize each object (vehicle V1) individually and also estimate the size of each object.

[0060] The tracking process predicts the next detection position from the time-series data of the object's (vehicle V1) position (distance and direction) and speed obtained from the clustering process, and identifies and tracks the object by comparing the actual detection position with the predicted position. The detection unit 24 further assigns a vehicle ID to each detected vehicle V1 in order to identify it. Note that this pre-processing may also be performed on the sensor unit 30 side.

[0061] The event detection process detects the occurrence of an event, the vehicle ID of the vehicle V1 involved in the event, and the location where the event occurred, based on the speed, position (lane, etc.), and driving conditions of each vehicle V1.

[0062] Specifically, the detection unit 24 detects that vehicle V1 is exceeding the speed limit by comparing the vehicle's speed with a predetermined speed threshold. The detection unit 24 also detects that vehicle V1 is traveling in the wrong direction by monitoring the vehicle's direction of travel for a certain period of time. Furthermore, the detection unit 24 detects that vehicle V1 is parked if its position does not change for a certain period of time (i.e., its speed is 0). In this case, the detection unit 24 detects that vehicle V1 is illegally parked depending on whether the parked location is a no-parking zone or not.

[0063] The detection unit 24 further detects the falling object M1 based on the object's speed, direction, and size. For example, if the object is smaller than a predetermined size (e.g., the size of a small vehicle) and stationary, the detection unit 24 recognizes the object as a falling object M1. Furthermore, for example, if the object is smaller than a predetermined size and is recognized as originating from behind a moving vehicle V1, the detection unit 24 recognizes the object as a falling object M1 from the vehicle V1.

[0064] The detection unit 24 further calculates, based on data from multiple vehicles, the number of vehicles V1 passing through each lane over a predetermined period of time (e.g., 5 to 10 minutes), the average speed of the vehicles V1, and the occupancy rate of the lanes by the vehicles V1, and detects congestion based on these calculation results.

[0065] When the detection unit 24 detects the occurrence of an event, it creates event information D2 related to the detected event. The event information D2 includes, for example, the type of event detected, the location (location information) where the event occurred, the time of occurrence, and the vehicle ID of the vehicle V1 involved in the event.

[0066] <Configuration of the control device> The management device 200, in terms of hardware configuration, has a control unit 201, a storage unit 202, and a communication unit 203, similar to the detection device 20. The control unit 201 includes a processing unit (processor) such as a CPU. The storage unit 202 includes a main storage unit and an auxiliary storage unit. The communication unit 203 functions as a communication interface.

[0067] <Software Configuration> Figures 4 and 5 are flowcharts showing the control structure of the program executed by the detection device 20.

[0068] Referring to Figure 4, this program includes the steps of receiving sensor information D1 from sensor 40 in step S201, executing a process to detect an event based on the received sensor information D1 in step S202, and branching the control flow according to the detected event in step S203. In step S202, in addition to the process to detect an event, a process to generate event information D2 related to the detected event is also executed. The events to be detected are events that may occur in areas A1 to A3, which are the target areas of sensor 40, and are, for example, events that are likely to cause traffic delays or accidents. The events to be detected are also events that are pre-set in the computer program stored in the memory unit 22. The events to be detected include, for example, the following events.

[0069] Speeding: An event representing speeding on the road by vehicle V1. Driving the wrong way: An event representing vehicle V1 driving the wrong way on road R1. Parking: An event indicating that vehicle V1 is parked on road R1. Falling Object: An event indicating that a falling object M1 is present on road R1. Traffic jam: An event indicating that traffic congestion is occurring on road R1.

[0070] This program is further executed when the detected event is "parking" or "falling object," and includes step S204 of referring to a selection table to select a camera 50 at the location where the event occurred, and step S205 of determining the shooting conditions for the selected camera.

[0071] This program is further executed when the detected event is "speeding" and includes steps S206 to select a camera 50 at the location where the event occurred by referring to a selection table, and step S207 to determine the shooting conditions of the selected camera.

[0072] This program is further executed when the detected event is "reverse driving" and includes step S208 of selecting a camera 50 at the event location by referring to a selection table, and step S209 of determining the shooting conditions of the selected camera.

[0073] This program is further executed when the detected event is "traffic jam" and includes steps S210 to select a camera 50 at the event location by referring to a selection table, and step S211 to determine the shooting conditions for the selected camera.

[0074] Referring to Figure 5, this program further includes step S214 of sending a control signal to the selected camera 50, and image Im transmitted from the camera 50 that sent the control signal. 1 Step S215 to receive the received image Im 1 The process includes the step S216 of detecting detailed event information D3, and the step S217 of storing the detected detailed information D3 in the storage unit 22 and transmitting it to the management device 200 via the communication unit 23 and the telecommunications network N1.

[0075] The detection device 20 repeatedly performs the above process.

[0076] Figure 6 is a flowchart showing the control structure of the program executed by the camera 50. Referring to Figure 6, this program consists of the steps of: taking a picture in normal mode (S301); receiving a control signal from the detection device 20 (S302); taking a picture in a predetermined shooting mode based on the instructions of the received control signal (S303); and capturing an image Im in the predetermined shooting mode. 1 The procedure includes step S304, which transmits the signal to the detection device 20 that transmitted the control signal. The normal mode in step S301 refers to a mode in which the entire view of the target area is captured using a number of frames that is less than or equal to the first number of frames F1.

[0077] <Detection System Operation> Figure 7 is a sequence diagram showing an example of a detection method performed by the detection system 10. The operation of the detection system 10 will be explained below, with reference to Figures 1 through 7 as appropriate.

[0078] Sensor 40a constantly emits electromagnetic waves to road R1 and receives reflected waves. Sensor 40a generates sensor information D1 (electrical signal) based on the received reflected waves and transmits the generated sensor information D1 to detection device 20a (step S1).

[0079] When the control unit 21 of the detection device 20a receives sensor information D1, it stores the received sensor information D1 in the storage unit 22. Based on the received sensor information D1, the detection unit 24 of the detection device 20a performs the above-mentioned preprocessing and event detection processing to detect the occurrence of a predetermined event, the vehicle ID of the vehicle V1 involved in the event, and the location (occurrence position) of the event, and creates event information D2 related to the detected event (step S2). The created event information D2 is stored in the storage unit 22. The event information D2 includes, for example, the type of event, the location of the event, the time of the event, the vehicle ID of the vehicle V1 related to the event, and the speed of the vehicle V1 related to the event.

[0080] The designated events may include events other than those listed above.

[0081] Next, the selection unit 25 extracts information regarding the type of event and the location where the event occurred from the event information D2. Depending on the type of event included in the event information D2, the selection unit 25 selects a camera 50 from among the multiple cameras 50a to 50c to be used to capture the image Im1 related to that event (Step S3: Second Step).

[0082] Next, the instruction unit 26 refers to the selection table and determines the shooting conditions for the selected camera 50 (step S4). The shooting conditions include, for example, the shooting location (center of road R1 or shoulder), zoom magnification, shooting start time, shooting time from start to finish, number of frames, etc.

[0083] For example, the selection unit 25 determines which type the detected event belongs to (step S203). If the event type is "parking" or "falling object", the selection unit 25 selects a camera 50 to photograph the location where the event occurred (steps S204, step S3). More specifically, based on the sensor information D1 from the sensor 40a, if a vehicle V1 parked in the first region A1, which is the target area of ​​the sensor 40a, the selection unit 25 selects a camera 50a to photograph the first region A1.

[0084] Next, the instruction unit 26 determines the shooting conditions for the selected camera 50a (steps S205 and S4). Specifically, the instruction unit 26 determines the shooting location and zoom magnification so that the license plate of the vehicle V1 is included. Also, since the parked vehicle V1 is not expected to move immediately (for example, within a few seconds), the instruction unit 26 determines the number of frames to a relatively small predetermined first number of frames F1 (for example, 5 frames per second) in order to save data capacity.

[0085] Furthermore, if a fallen object M1 in the first region A1 is detected based on the sensor information D1 from sensor 40a, the selection unit 25 selects a camera 50a to photograph the first region A1 (step S204). The instruction unit 26 then determines the shooting location so that the location of the fallen object M1 is included, and determines the zoom magnification according to the size of the fallen object M1. Also, since the fallen object M1 is not expected to move immediately, similar to a parked vehicle V1, the instruction unit 26 determines the number of frames to be the first number of frames F1 (step S205).

[0086] If a fallen object M1 is detected, it must be removed. The details of the removal work will vary depending on the type of object M1 (for example, whether it is a heavy object or not) and location (for example, whether it has fallen in the middle of road R1 or on the shoulder of road R1). The workers performing the removal work will make a judgment based on the details of the fallen object M1 described below (D3) and proceed with the removal work.

[0087] Therefore, when a fallen object M1 is detected, the instruction unit 26 may determine both the shooting conditions for identifying the object M1 and the shooting conditions for identifying the location of the fallen object M1. The shooting conditions for identifying the object are, for example, conditions for zooming in and shooting the fallen object M1 in order to identify in detail what the fallen object M1 is. The shooting conditions for identifying the location are, for example, conditions for shooting the entire view of the first area A1 including the fallen object M1 in order to identify in detail where the fallen object M1 is located on the road R1. For example, the instruction unit 26 may instruct the camera 50a to perform shooting for object identification for a predetermined shooting time, and then perform shooting for location identification for a predetermined shooting time as the shooting conditions.

[0088] Furthermore, if the vehicle V1 that dropped the object M1 is also detected when detecting the fallen object M1, the selection unit 25 may select a camera 50 that photographs a location downstream from the location where the event occurred (the location of the fallen object M1), and the instruction unit 26 may determine the shooting location and zoom magnification of the camera 50 so that the license plate of the vehicle V1 is included.

[0089] If the event is "speeding," the selection unit 25 selects a camera 50 that photographs the location where the event occurred and a camera 50 that photographs a location downstream from the location where the event occurred (steps S206, S3).

[0090] More specifically, based on sensor information D1 from sensor 40a, if a vehicle V1 traveling in the first region A1 at a speed exceeding a predetermined speed is detected, the selection unit 25 selects camera 50a to photograph the first region A1 and cameras 50b and 50c to photograph the area downstream of the first region A1. Alternatively, the selection unit 25 may not select a camera 50 to photograph the location where the event occurred, but instead select only cameras 50 to photograph locations downstream of the event location.

[0091] Next, the instruction unit 26 determines the shooting conditions for the selected cameras 50a, 50b, and 50c (steps S207 and S4). Specifically, the instruction unit 26 determines the shooting time for cameras 50a, 50b, and 50c based on the event time included in the event information D2 and the speed of the vehicle V1. The instruction unit 26 also determines the shooting location and zoom magnification for cameras 50a, 50b, and 50c so that the license plate of the vehicle V1 is included in the image.

[0092] Furthermore, in order to more reliably photograph the license plate of vehicle V1 traveling at a predetermined speed, the instruction unit 26 sets the number of frames to a second number of frames F2 (for example, 30 frames per second), which is greater than the first number of frames F1. The number of frames may also be determined based on the speed of vehicle V1. For example, the faster the speed of vehicle V1, the more frames may be used.

[0093] If the event is "reverse travel", the selection unit 25 selects a camera 50 that photographs the location where the event occurred and a camera 50 that photographs a location upstream of the location where the event occurred (steps S208, S3).

[0094] More specifically, if a vehicle V1 traveling in the second region A2 in the opposite direction to the direction of travel AR1 is detected based on sensor information D1 from sensor 40b, the selection unit 25 selects camera 50b to photograph the second region A2 and camera 50a to photograph the area upstream of the second region A2. Alternatively, the selection unit 25 may not select a camera 50 to photograph the location where the event occurred, but only select a camera 50 to photograph the area upstream of the event location.

[0095] Next, the instruction unit 26 determines the shooting conditions for the selected cameras 50a and 50b (steps S209 and S4). Specifically, the instruction unit 26 determines the shooting time for cameras 50a and 50b based on the time of the event included in the event information D2 and the speed of the vehicle V1. The instruction unit 26 also determines the shooting location and zoom magnification for cameras 50a and 50b so that the license plate of the vehicle V1 is included. Furthermore, in order to more reliably capture the license plate of the moving vehicle V1, the instruction unit 26 determines the number of frames to be a second number of frames F2, which is greater than the first number of frames F1.

[0096] If the event is "traffic congestion," the selection unit 25 selects a camera 50 to photograph the location where the event occurred (steps S210, S3). More specifically, if traffic congestion is detected in the first region A1 based on the sensor information D1 from the sensor 40a, the selection unit 25 selects a camera 50a to photograph the first region A1.

[0097] Furthermore, in order to continuously monitor the starting point (downstream end) and ending point (upstream end) of the traffic congestion, the selection unit 25 may further select cameras 50 that photograph locations upstream and downstream of the event occurrence location.

[0098] Next, the instruction unit 26 determines the shooting conditions for the selected camera 50a (steps S211 and S4). Specifically, the instruction unit 26 determines the zoom magnification of the camera 50a (e.g., 1x) so that the entire view of the first region A1 is included. Also, since the vehicle V1 included in the traffic jam is traveling at a relatively low speed and the traffic jam situation is not expected to change immediately (e.g., within a few seconds), the instruction unit 26 determines the number of frames to be the first number of frames F1.

[0099] Next, the instruction unit 26 instructs the camera 50 selected by the selection unit 25 to take a picture (steps S5 to S7). For example, if camera 50a (or camera 50b) is selected, the instruction unit 26 of the detection device 20a transmits a control signal to camera 50a (or camera 50b) (steps S214, S5). Also, if camera 50c is selected, the instruction unit 26 of the detection device 20a transmits a control signal to the detection device 20b that controls camera 50c via the telecommunications network N1 (steps S214, S6). Then, the detection device 20b transmits a control signal to camera 50c (step S7).

[0100] Camera 50 normally operates in normal mode (steps S301, S8, S9). Normal mode refers to a mode in which the entire area of ​​the target is captured using, for example, a number of frames equal to or less than the first frame number F1. Camera 50 may also normally operate in standby mode (a mode in which it does not take pictures and is on standby in a power-saving manner).

[0101] When the camera 50 receives a control signal from the instruction unit 26 (step S302), the camera 50 operates in a predetermined shooting mode based on the control signal (steps S303, S10, S11). The predetermined shooting mode refers to a mode in which the camera shoots according to the various shooting conditions determined by the instruction unit 26 in step S4.

[0102] When camera 50 finishes shooting in shooting mode, it transmits image Im1 to detection device 20 (steps S304, S12-S14). Detection device 20 stores the received image Im1 in storage unit 22. Specifically, cameras 50a and 50b transmit image Im1 to detection device 20a (step S12). Camera 50c also transmits image Im1 to detection device 20b (step S13), and detection device 20b transmits image Im1 to detection device 20a via telecommunications network N1 (step S14). The control unit 21 of detection device 20a stores image Im 1 The received image Im 1 The data is stored in the memory unit 22.

[0103] Next, the detailed detection unit 27 of the detection device 20a detects detailed event information D3 based on the event information D2 and the image Im1 (steps S216, S15). For example, if the event is "falling object", the detailed detection unit 27 crops the area where the falling object M1 is visible from the image Im1 based on the event information D2 and detects the cropped image as detailed information D3. Alternatively, the detailed detection unit 27 may detect the image Im1 itself as detailed information D3 without cropping the image Im1.

[0104] Furthermore, if the event type is "parking," "speeding," or "driving the wrong way," the detailed detection unit 27 identifies the location of the vehicle V1's license plate in image Im1 based on the event information D2. The detailed detection unit 27 then reads the characters on the license plate and detects this character information as detailed information D3. The detailed detection unit 27 may also detect a cropped image with the license plate portion cropped as detailed information D3. In other words, the detailed detection unit 27 detects information related to the vehicle V1's license plate (information including the character information of the license plate and at least one image containing the license plate) as detailed information D3. Also, if the event type is "traffic jam," the detailed detection unit 27 detects image Im1 itself as detailed information D3.

[0105] The detailed detection unit 27 stores the detected detailed information D3 in the storage unit 22 and transmits the detailed information D3 to the management device 200 via the communication unit 23 and the telecommunications network N1 (steps S217, S16). The control unit 201 of the management device 200 stores the detailed information D3 received by the communication unit 203 in the storage unit 202.

[0106] <Effects of this embodiment> The detection device 20 includes a selection unit 25 that selects a camera 50 from among multiple cameras 50 installed on the road R1 to be used to capture an image Im1 related to the detected event, and an instruction unit 26 that instructs the selected camera 50 to take a picture. Therefore, a more suitable image Im1 can be recorded according to the detected event. In addition, detailed event information D3 can be detected more accurately based on the image Im1.

[0107] For example, if the event type is "speeding," a camera 50 located downstream from where the event was detected is instructed to take a picture, thus more reliably capturing the moving vehicle V1 in image Im1. Similarly, if the event type is "driving in the wrong direction," a camera 50 located upstream from where the event was detected is instructed to take a picture, thus more reliably capturing the moving vehicle V1 in image Im1.

[0108] In particular, the instruction unit 26 determines the shooting conditions for the camera 50 selected by the selection unit 25 according to the detected event, and instructs the camera 50 selected by the selection unit 25 to take a picture according to those shooting conditions. As a result, a more suitable image Im1 can be obtained according to the event, and detailed event information D3 can be detected more accurately based on the image Im1.

[0109] For example, if the event is "speeding" or "driving the wrong way," the instruction unit 26 determines the number of frames for the selected camera 50 to be a second number of frames F2, which is greater than the first number of frames F1. This makes it possible to more reliably include the moving vehicle V1 in the image Im1. Also, if the event is "parking," "speeding," or "driving the wrong way," the shooting location and zoom magnification of the selected camera 50 are determined so that the license plate of the vehicle V1 is captured, thus enabling more accurate detection of detailed information D3, which includes information about the license plate.

[0110] <Variations> The following describes modified examples of the embodiments. In the modified examples, parts that remain unchanged from the embodiments are denoted by the same reference numerals and their descriptions are omitted.

[0111] <Event detection using machine learning> The detection unit 24 may be configured to detect one or more events that occurred on road R1 from among a set of events, using a learning model trained by machine learning.

[0112] Figure 8 is a block diagram illustrating the processing performed by the trained classification model. The memory unit 22 stores the trained discrimination model MD1. The discrimination model MD1 is a model that has been trained using a predetermined learning algorithm LA1 to recognize the correspondence between multiple types of events and labels L1, for example, using training data LD1 (supervisory data). The learning algorithm LA1 can be, for example, a support vector machine. The learning algorithm LA1 may also be a different algorithm from a support vector machine (for example, a neural network such as deep learning).

[0113] In this modified version, the object's feature vector FV1 is extracted by preprocessing the input sensor information D1. In this preprocessing, signal processing is used to extract the feature vector FV1 that is effective for event detection from the sensor information D1. The extracted feature vector FV1 is input to the discrimination model MD1, and the label L1, which is the event detection result, is output.

[0114] Figure 9 is a block diagram illustrating the process of generating the training data LD1. The training data LD1 is generated by individually detecting and labeling each event. Events such as driving in the wrong direction, speeding, and traffic congestion can be automatically detected from sensor information D1 as described above. When these events are detected, data within a predetermined time range including the event detection time is extracted, and the training data LD1 can be generated by associating the label L1 of each event with the extracted data.

[0115] On the other hand, it is preferable to manually generate the training data LD1 related to parking (illegal parking) and fallen objects. Specifically, for example, the sensor 40 detects various illegal parkings and various fallen objects within the target area of ​​the sensor 40, and the operator generates the training data LD1 by inputting the corresponding label L1 based on the sensor information D1 displayed on the display. By creating an identification model MD1 using such training data LD1, it becomes possible to accurately detect multiple types of events. In particular, the detection accuracy of events such as parked vehicles and fallen objects can be improved.

[0116] <Examples of cases where control signals conflict> In the above embodiment, an event is detected based on sensor information D1, and in step S5, for example, a control signal including one shooting condition is sent to the camera 50. However, in reality, multiple events may occur simultaneously on the road R1. For example, while a fallen object M1 is present in the first area A1, a vehicle V1 driving in the wrong direction may occur in the second area A2.

[0117] In this case, the detection unit 24 of the detection device 20a determines that an event called "falling object" has occurred based on the sensor information D1 from sensor 40a, and also determines that an event called "reverse driving" has occurred based on the sensor information D1 from sensor 40b. The selection unit 25 selects a camera 50a to photograph the location where the "falling object" occurred, according to the detected event "falling object," and the instruction unit 26 determines the shooting conditions (for example, setting the zoom magnification to 1x and the number of frames to the first frame number F1 in order to photograph the entire view of the first region A1). The instruction unit 26 then transmits a control signal CS1 corresponding to the "falling object" to the camera 50a.

[0118] Furthermore, the selection unit 25 selects a camera 50a that photographs an area upstream of the location where the "wrong-way driving" event occurred, and the instruction unit 26 determines the shooting conditions (for example, a zoom magnification greater than 1x and a frame count of 2F2 in order to photograph the license plate of vehicle V1). The instruction unit 26 then transmits a control signal CS2 corresponding to "wrong-way driving" to the camera 50a.

[0119] Thus, when the detection system 10 detects multiple events occurring on road R1 at the same time, multiple control signals CS1 and CS2 may be transmitted to the camera 50 at the same time. In other words, multiple control signals CS1 and CS2 may conflict with each other in a single camera 50.

[0120] In this case, it is conceivable that camera 50 would take pictures in the order in which the control signals are input. However, if, for example, control signal CS1 is input to camera 50a first, and camera 50a takes a picture of the entire first area A1 for a predetermined shooting time based on control signal CS1, there is a risk that a vehicle V1 moving in the wrong direction may pass through the first area A1 during that shooting. In this case, there is a risk that the vehicle V1 moving in the wrong direction may be missed.

[0121] Therefore, in this modified version, a priority parameter is added to the control signal for each type of event. For example, if the event type is "speeding," the target of the camera is a moving vehicle V1, and since vehicle V1 is likely to decelerate and escape the state of exceeding a predetermined speed, the time during which camera 50 can photograph vehicle V1 while the event is occurring is limited. For this reason, the priority for photographing "speeding" is set to be the highest.

[0122] Furthermore, when the event type is "driving in the wrong direction," the subject of the filming is a moving vehicle V1, so the time that camera 50 can film vehicle V1 while the event is occurring is somewhat limited. However, compared to the case of "speeding," vehicle V1 is less likely to escape the state of driving in the wrong direction, so even if camera 50c misses filming vehicle V1 driving in the wrong direction, for example, there is a high possibility that it can be filmed by another camera 50a. For this reason, the priority of filming "driving in the wrong direction" is set lower than that of "speeding."

[0123] Furthermore, when the event type is "parking," the subject of the filming is a parked vehicle V1, so the time during which camera 50 can film vehicle V1 while the event is occurring is longer compared to when the event type is "speeding" or "driving the wrong way." On the other hand, since a parked vehicle V1 may start moving and leave its location, it is preferable to film it earlier than when the event is "falling object." For this reason, the priority of filming related to "parking" is set lower than for "speeding" and "driving the wrong way," and higher than for "falling object."

[0124] Furthermore, when the event type is "traffic jam," there is no need to obtain license plate information based on the image or identify fallen objects, for example, so the need for images is lower compared to other events. For this reason, the priority of images related to "traffic jam" is lower than that of other events. Based on the above, the priority of each event type in this modified example, in descending order, is speeding, driving the wrong way, parking, fallen objects, and traffic jam. Note that this priority is just an example, and other order is also acceptable.

[0125] Furthermore, if multiple control signals conflict in a single camera 50, the camera will capture images in order of the control signals corresponding to the highest priority event. For example, if camera 50a receives a control signal CS1 corresponding to "falling object," and then camera 50a receives a control signal CS2 corresponding to "driving in the wrong direction" while camera 50a is capturing images of the falling object M1, camera 50a will temporarily interrupt the capture based on control signal CS1 and capture images of the vehicle V1 driving in the wrong direction based on the higher priority control signal CS2. This configuration allows for more favorable image capture even when multiple control signals conflict.

[0126] The detection device 20 according to the above embodiment is provided separately from the sensor unit 30. However, part or all of the detection device 20 may be included in the sensor unit 30. For example, a computer may be mounted on the sensor unit 30, and the computer may detect events based on the sensor information D1 of the sensor 40. In this case, the sensor unit 30 to The onboard computer functions as a detection unit 24.

[0127] In other words, the detection device 20 may be implemented by a computer installed in one location, as in the embodiment described above, or by a plurality of computers distributed among the sensor unit 30.

[0128] <Examples of camera and sensor variations> In the above embodiment, since the sensor unit 30 is equipped with a sensor 40 and a camera 50, there is a one-to-one correspondence between the sensor 40 and the camera 50, and the installation distance between the sensor 40 and the camera 50 is the same. However, the sensor 40 and the camera 50 may correspond to many sensors, and the installation distance between the sensor 40 and the camera 50 may be different.

[0129] For example, when using a sensor 40 that can monitor an area of ​​200m and a camera 50 that can monitor an area of ​​100m, in order to detect events in the first area A1 of 200m, one sensor 40 may be associated with two cameras 50, with the sensors 40 installed every 200m and the cameras 50 installed every 100m.

[0130] <Modification of the detection unit> The detection system in this modified example has a function in which multiple sensor units 30 (sensors 40) and multiple detection devices 20 work in coordination. This allows the system to track a vehicle V1 traveling across the target area of ​​sensor 40. In this modified example, events such as speeding or driving in the wrong direction are assumed to be the events to be detected. That is, when the detection system 10 detects an event such as speeding or driving in the wrong direction, it identifies the vehicle V1 that is the target of the event and tracks the identified event target vehicle V1 beyond the target area where the event was detected. Furthermore, the detection system 10 records the event target vehicle V1 while tracking it by switching the selection of cameras 50 that photograph the event target vehicle V1 according to the tracking situation.

[0131] Multiple sensor units 30 operate in coordination by operating based on the same time. Each of the multiple sensor units 30 synchronizes its time by, for example, obtaining time information from an NTP (Network Timing Protocol) server.

[0132] Figure 10 is a flowchart showing the sequence of operations performed by the detection devices 20a and 20b according to this modified example. This example describes the processing of a portion of the tracking section when tracking the event target vehicle V1. Hereafter, for the purpose of distinction, the sensor information D1 acquired from sensors 40a and 40c will be referred to as sensor information D1a and D1c, respectively, and the event information D2 detected based on sensors 40a and 40c will be referred to as event information D2a and D2c, respectively.

[0133] Referring to Figure 1, for example, suppose that vehicle V1 is traveling at a predetermined speed in the first region A1. The detection device 20a detects that vehicle V1 is exceeding the speed limit. Specifically, the detection device 20a receives sensor information D1a from sensor 40a (step S401). Subsequently, the detection unit 24 of the detection device 20a detects the event "speeding" based on the received sensor information D1a and generates event information D2a including the vehicle ID, position, speed, and size of vehicle V1 (step S402). Depending on the detected event (speeding), the detection device 20a selects a camera 50 to photograph the location where the event occurred and a camera 50 to photograph a location downstream of the location where the event occurred. The detection device 20a issues a shooting instruction to the camera 50 photographing the location where the event occurred and transmits the event information D2a to the detection device 20b located downstream (step S403).

[0134] The detection device 20b receives sensor information D1c from sensor 40c (step S501). The detection device 20b receives event information D2a from detection device 20a (step S502). Note that the detection device 20b may receive sensor information D1c after receiving event information D2a. Based on the event information D2a, the detection device 20b extracts vehicle V1 information from sensor information D1c (step S503). With this configuration, even if the sensor information D1c acquired from sensor 40c does not include the event "speeding", vehicle V1 information (e.g., position, speed) can be obtained from the sensor information D1c.

[0135] The detection device 20b further assigns the same ID (or corresponding ID) as the vehicle ID included in the event information D2a received from the detection device 20a to the event information D2c generated based on the sensor 40c. This makes it possible to link the event information D2a detected based on the sensor 40a with the event information D2c detected based on the sensor 40c. Since the same (or corresponding) ID is assigned to the vehicle V1 in the separate event information D2a and D2c, it becomes easier to track the vehicle V1.

[0136] The detection device 20b, upon detecting vehicle V1, selects a camera 50 to photograph vehicle V1 and determines the shooting conditions. The detection device 20b issues a shooting instruction to the selected camera and also receives event information D2a from the detection device 20a and the event information it itself has detected. Event Information D2c is transmitted to another detection device located downstream of detection device 20b. In this way, the detection system according to this modified example tracks and records the speeding vehicle V1.

[0137] While this modification illustrates an example of detecting a speeding event, the disclosure is not limited to such examples. For instance, if a wrong-way driving event is detected, the system may track and record the event vehicle. In this case, the event information is transmitted to another detection device located upstream of the detection device that detected the event.

[0138] "others" In the above embodiment, the sensor 40 transmits electromagnetic waves to the road R1 and acquires sensor information D1, which includes information about events occurring on the road R1, based on the reflected waves. However, the sensor 40 may also transmit electromagnetic waves to an area other than the road R1 and acquire sensor information D1, which includes information about events occurring in that area. For example, if debris M1 is attached to a slope beside the road R1, the debris M1 may be moved by wind or other factors and enter the road R1. For this reason, the sensor 40 may acquire sensor information D1 not only from the road R1 but also from an area located near the road R1. Furthermore, the detection device 20 may detect events in the area located near the road R1 that may interfere with the passage of vehicles V1 on the road R1 in the future.

[0139] In the above embodiment, based on the sensor information D1, at least one event is detected from among a predetermined set of multiple types of events. However, the predetermined set of events does not necessarily have to be of multiple types; it may be just one type. Even in this case, when the detection unit 24 detects a predetermined event, it selects a camera 50 from among the multiple cameras 50 installed on the road R1 to capture images related to the event, according to the content of the event. Examples of event content include the location where the event occurred and the type of event. For example, the detection unit 24 selects a camera 50 suitable for capturing the event (for example, a camera 50 close to the event location) according to the content of the detected event (i.e., the location where the event occurred).

[0140] Alternatively, for example, the detection unit 24 may detect only "speeding" as an event based on the sensor information D1. 4 Only the route S203→S206→S207 may be selected. In this case, the selection unit 25 selects a camera 50 that photographs the location where the event occurred and a camera 50 that photographs a location downstream of the location where the event occurred, thereby preventing the vehicle V1 from being missed and allowing for more accurate recording of image information related to the event (speeding).

[0141] Furthermore, for example, the detection unit 24 may detect only "driving in the wrong direction" as an event based on the sensor information D1. That is, Figure 4 Only the route S203→S208→S209 may be selected. In this case, the selection unit 25 selects a camera 50 that photographs the location where the event occurred and a camera 50 that photographs a location upstream of the location where the event occurred, thereby preventing the vehicle V1 from being missed and allowing for more accurate recording of image information related to the event (driving in the wrong direction).

[0142] [Additional Note] Furthermore, at least some of the embodiments and various modifications described above may be combined in any way. Also, the embodiments disclosed herein should be considered in all respects to be illustrative and not restrictive. The scope of this disclosure is indicated by the claims, and all modifications within the meaning and scope of the claims are intended to be included. [Explanation of symbols]

[0143] 10 Detection Systems 20, 20a, 20b detection device 21 Control Unit 22 Memory section 23 Communications Department 24 Detection unit 25 Selection Section 26 Instruction section 27 Detailed detection unit 200 Management device 201 Control Unit 202 Storage section 203 Communications Department 30, 30a, 30b, 30c Sensor Unit 31,31a Enclosure 40, 40a, 40b, 40c sensors 50, 50a, 50b, 50c Camera 51 Moving parts 52 Zoom Lens 53 imaging element 6a,6b Post TC1 Traffic Control Center N1 Telecommunications Network R1 Road A1 1st area A2 2nd area A3 3rd area V1 Vehicle M1 Fallen object AR1 Traffic direction D1, D1a, D1c Sensor Information D2, D2a, D2c Event Information D3 Detailed Information Im1 Image F1 First Frame F2 Second frame CS1, CS2 control signals FV1 Features L1 Label LD1 training data LA1 learning algorithm MD1 identification model

Claims

1. A detection unit that obtains sensor information from a sensor that detects an object by transmitting electromagnetic waves to the road and receiving the electromagnetic waves reflected by the object, and based on the obtained sensor information, detects one or more events from a preset set of multiple types of events, A selection unit selects a camera from among multiple cameras installed on the road to capture images related to the event, according to the content of the event detected by the detection unit. An instruction unit transmits a control signal to the camera selected by the selection unit to instruct it to take a picture, Equipped with, The control signal is assigned a priority corresponding to the type of event. The aforementioned priority is information used to instruct the camera on the order in which to take pictures based on multiple control signals when multiple control signals conflict. Detection device.

2. The detection device according to claim 1, wherein the aforementioned set plurality of events include events that may occur in the target area where the sensor acquires the sensor information.

3. The aforementioned pre-configured events are Driving a vehicle at a speed exceeding the legal or designated speed limit, Vehicles driving against traffic on the road, Parking vehicles on the road, Road congestion, and The presence of fallen objects on the road, including at least one of the above, The detection device according to claim 1 or claim 2.

4. When the detection unit detects speeding on the road as the event, the selection unit selects a camera from among the plurality of cameras that captures an area downstream of the area where the sensor acquires sensor information, relative to the direction of travel on the road. The detection device according to claim 1 or claim 2.

5. When the detection unit detects a vehicle driving in the wrong direction on the road as the event, the selection unit selects from among the plurality of cameras a camera that captures an area upstream of the direction of travel on the road, relative to the area where the sensor acquires sensor information. The detection device according to claim 1 or claim 2.

6. The instruction unit, in response to the event detected by the detection unit, determines either a first shooting condition for shooting at a predetermined frame rate, or a second shooting condition for shooting at a frame rate higher than the predetermined frame rate, as the shooting condition for the camera selected by the selection unit, and issues an instruction to shoot according to the determined shooting condition. The detection device according to claim 1 or claim 2.

7. When the detection unit detects, as a preset event, a vehicle parked on the road, traffic congestion on the road, or the presence of an object on the road, the instruction unit determines the first shooting conditions as the shooting conditions for the camera selected by the selection unit, and issues an instruction to shoot according to the determined first shooting conditions. The detection device according to claim 6, wherein when the detection unit detects, as a preset event, a vehicle speeding on the road exceeding the legal speed or designated speed, or a vehicle driving in the wrong direction on the road, the instruction unit determines the second shooting conditions as the shooting conditions for the camera selected by the selection unit, and gives an instruction to take a photograph according to the determined second shooting conditions.

8. The system further includes a detailed detection unit that detects detailed information about an event detected by the detection unit based on an image captured by the camera selected by the selection unit. The detection device according to claim 1 or claim 2.

9. If the detection unit detects, as a preset event, a vehicle exceeding the legal speed limit or designated speed, driving the wrong way on a road, or parking a vehicle on a road, The detection device according to claim 8, wherein the detailed detection unit detects information regarding the license plate of the target vehicle as the detailed information.

10. The aforementioned sensor and, Multiple cameras, A detection device according to claim 1 or claim 2, A detection system equipped with the following features.

11. A computer program to be executed by a computer, The aforementioned computer program, The process involves obtaining sensor information from a sensor that detects an object by transmitting electromagnetic waves to a road and receiving the electromagnetic waves reflected by the object, and then detecting one or more events from a preset set of multiple types of events based on the acquired sensor information. Depending on the content of the detected event, the process involves selecting a camera from among several cameras installed on the road to capture images related to the event, The steps include: sending a control signal to the selected camera to instruct it to take a picture; Includes, The control signal is assigned a priority corresponding to the type of event. The aforementioned priority is information used to instruct the camera on the order in which to take pictures based on multiple control signals when multiple control signals conflict. Computer program.

Images