Method and device for determining vehicle attribute information, medium and program product

Abstract

The invention relates to the technical field of image processing, in particular to a method and device for determining vehicle attribute information, a medium and a program product, and the method comprises the steps: receiving alarm information sent by a target edge computing device when determining that a vehicle triggers a related event; determining joint identification information of the vehicle based on attribute information identification carried in the alarm information; and querying attribute information of the vehicle from other edge computing devices except the target edge computing device based on the joint identification information. Through the method, more comprehensive attribute information of the vehicle can be obtained.

Description

Technical Field

[0001] The present application relates to the field of image processing technology, and in particular to a method, device, medium and program product for determining vehicle attribute information. Background Art

[0002] In drop-and-hook transport, the tractor and trailer can be separated. This innovative model greatly improves transportation efficiency. Through drop-and-hook transport, when the tractor arrives at the destination, it can quickly unload the fully loaded trailer and hang on the empty trailer, thus saving the time of loading and unloading goods and realizing fast and efficient transportation of goods. In addition to the above-mentioned large vehicle towing trailers, small household vehicles can also hang a trailer board behind them for convenient travel.

[0003] In transportation, trailers are usually equipped with two license plates, one for the front of the vehicle (tractor) and the other for the trailer (trailer). Since intelligent traffic cameras determine whether a vehicle is violating traffic rules by photographing the area near the "standard line" of the intersection, when the above-mentioned types of vehicles violate traffic rules, the license plate information will be captured. However, since the rear of the vehicle is facing the camera at this time, the license plate captured is the trailer plate or trailer license plate, but the tractor actually caused the violation. At this time, the license plate information of the actual violating vehicle cannot be accurately obtained. Summary of the invention

[0004] The embodiments of the present application provide a method, device, medium and program product for determining vehicle attribute information, which can obtain more comprehensive vehicle attribute information.

[0005] In a first aspect, an embodiment of the present application provides a method for determining vehicle attribute information, which is applied to a server, and the method includes: Receive warning information sent by the target edge computing device when determining that the vehicle triggers a related event; Determine the joint identification information of the vehicle based on the attribute information identifier carried in the warning information, wherein the joint identification information includes attribute information identifiers set by different edge computing devices for the vehicle, the attribute information of the vehicle recorded by the different edge computing devices is different, and the attribute information recorded in the edge computing device is obtained through a camera; Based on the joint identification information, query the attribute information of the vehicle from other edge computing devices except the target edge computing device.

[0006] In an embodiment of the present application, based on the attribute information identifier sent by the target edge computing device, the joint identification information of the vehicle is determined, and based on the joint identification information, the attribute information of the vehicle in each edge computing device is located to obtain the full amount of attribute information of the vehicle, thereby providing a complete evidence trajectory chain for relevant personnel.

[0007] In a possible implementation, the joint identification information of the vehicle is determined in the following manner: For any two edge computing devices, receiving a first bird's-eye view including a first vehicle from one of the any two edge computing devices, and receiving a second bird's-eye view including a second vehicle from another edge computing device of the any two edge computing devices, wherein the bird's-eye view is obtained by converting images captured by cameras corresponding to the edge computing devices, and the vehicle travels from a first capturing area of ​​a first camera corresponding to the one edge computing device to a second capturing area of ​​a second camera corresponding to the other edge computing device; When the first vehicle and the second vehicle are determined to be the same vehicle based on the generation time of the first bird's-eye view, the coordinates of the first vehicle in the first bird's-eye view, the generation time of the second bird's-eye view, and the coordinates of the second vehicle in the second bird's-eye view, the joint identification information of the vehicles is constructed using the attribute information identifier of the first vehicle obtained from the one edge computing device and the attribute information identifier of the second vehicle obtained from the other edge computing device, wherein the coordinate systems of the bird's-eye views corresponding to different edge computing devices are the same.

[0008] In the above embodiment, based on the feature that the coordinate systems of the bird's-eye views corresponding to different edge computing devices are the same, it is determined whether the vehicles in the bird's-eye views corresponding to different edge computing devices are the same vehicle through the generation time of the bird's-eye view and the position of the vehicle in the bird's-eye view, and the attribute information identifiers of the same vehicle in different edge computing devices are associated to obtain joint identification information, so as to obtain accurate and comprehensive attribute information of the vehicle.

[0009] In a possible implementation, determining that the first vehicle and the second vehicle are the same vehicle based on the generation time of the first bird's-eye view, the coordinates of the first vehicle in the first bird's-eye view, the generation time of the second bird's-eye view, and the coordinates of the second vehicle in the second bird's-eye view includes: When the generation time of the first bird's-eye view matches the generation time of the second bird's-eye view, and the coordinates of the first vehicle in the first bird's-eye view match the coordinates of the second vehicle in the second bird's-eye view, it is determined that the first vehicle and the second vehicle are the same vehicle.

[0010] In a possible implementation, determining that the first vehicle and the second vehicle are the same vehicle based on the generation time of the first bird's-eye view, the coordinates of the first vehicle in the first bird's-eye view, the generation time of the second bird's-eye view, and the coordinates of the second vehicle in the second bird's-eye view includes: Predicting the time when the first vehicle appears in the second shooting area based on the generation time of the first bird's-eye view, and predicting the coordinates of the first vehicle when it appears in the second shooting area based on the coordinates of the first vehicle in the first bird's-eye view; When the time of the appearance matches the generation time of the second bird's-eye view, and the coordinates of the first vehicle when it appears in the second shooting area match the coordinates of the second vehicle in the second bird's-eye view, the first vehicle and the second vehicle are the same vehicle.

[0011] In the above embodiment, two solutions are provided for judging whether the vehicles are the same vehicle based on the generation time of the bird's-eye view and the position of the vehicle in the bird's-eye view. Different solutions are adopted according to different scenarios to obtain more accurate results.

[0012] In a possible implementation manner, the attribute information of the vehicle includes at least one of the following: The rear license plate information of the vehicle is obtained from the edge computing device corresponding to the camera for collecting the rear license plate information; The front license plate information of the vehicle is obtained from the edge computing device corresponding to the camera used to collect the front license plate information; The side information of the vehicle is obtained from the edge computing device corresponding to the camera used to collect side information.

[0013] In the above embodiment, multiple cameras are provided to respectively collect different attribute information of the vehicle, so as to obtain more comprehensive information and provide complete evidence for relevant personnel.

[0014] In a second aspect, an embodiment of the present application provides a method for determining vehicle attribute information, which is applied to an edge computing device, and the method includes: Acquire image data of the vehicle through a camera; When it is determined based on the image data that the vehicle triggers a related event, an alarm message is sent to a server, so that the server determines the joint identification information of the vehicle based on the attribute information identifier carried in the alarm message, wherein the joint identification information includes the attribute information identifier set for the vehicle by different edge computing devices, and the attribute information of the vehicle recorded by the different edge computing devices is different; Receive the joint identification information sent by the server, and send the attribute information of the vehicle corresponding to the joint identification information to the server.

[0015] In an embodiment of the present application, the target edge computing device can determine whether the vehicle triggers a related event, and then can send an alarm message to the server in a timely manner so that relevant personnel can handle the vehicle that triggers the related event; the edge computing server can also obtain the vehicle's attribute information locally based on the joint identification information and send it to the server, so that the server can obtain accurate attribute information.

[0016] In a possible implementation manner, before determining the vehicle triggering related event based on the image data, the method further includes: Based on the image data, the vehicle is determined to be a target type of vehicle through a pre-constructed recognition model, and an attribute information identifier is generated for the target type of vehicle. The recognition model is trained based on sample images corresponding to the target type of vehicle and the vehicle type corresponding to the sample images.

[0017] In the above embodiment, the target type of vehicle can be identified through the pre-built recognition model, and then the attribute information of the target type of vehicle can be obtained.

[0018] In a third aspect, an embodiment of the present application provides a device for determining vehicle attribute information, which is applied to a server, and the device includes: A receiving module, used to receive the warning information sent by the target edge computing device when determining that the vehicle triggers a related event; an identification determination module, configured to determine joint identification information of the vehicle based on the attribute information identification carried in the warning information, wherein the joint identification information includes attribute information identifications set by different edge computing devices for the vehicle, the attribute information of the vehicle recorded by the different edge computing devices is different, and the attribute information recorded in the edge computing device is obtained through a camera; A query module is used to query the attribute information of the vehicle from other edge computing devices except the target edge computing device based on the joint identification information.

[0019] In a fourth aspect, an embodiment of the present application provides a device for determining vehicle attribute information, which is applied to an edge computing device, and the device includes: An acquisition module, used for acquiring image data of the vehicle through a camera; a sending module, configured to send an alarm message to a server when it is determined based on the image data that the vehicle triggers a related event, so that the server determines the joint identification information of the vehicle based on the attribute information identifier carried in the alarm message, wherein the joint identification information includes the attribute information identifier set for the vehicle by different edge computing devices, and the attribute information of the vehicle recorded by different edge computing devices is different; The receiving module is used to receive the joint identification information sent by the server, and send the attribute information of the vehicle corresponding to the joint identification information to the server through the sending module.

[0020] In a fifth aspect, an embodiment of the present application provides a system for determining vehicle attribute information, the system comprising: a plurality of cameras, an edge computing device end, and a server end; The camera is used to obtain image data of the vehicle and send the obtained image data to the corresponding edge computing device end, and the camera corresponds to the edge computing device end one by one; The edge computing device is used to receive image data sent by the corresponding camera, and when it is determined based on the image data that the vehicle triggers a related event, send an alarm message to the server; and receive joint identification information sent by the server, and send attribute information of the vehicle corresponding to the joint identification information to the server; The server is used to receive the warning information sent by the target edge computing device when determining that the vehicle triggers a related event; determine the joint identification information of the vehicle based on the attribute information identifier carried in the warning information, and query the attribute information of the vehicle from other edge computing devices except the target edge computing device; The joint identification information includes attribute information identifiers set by different edge computing devices for the vehicle, and the attribute information of the vehicle recorded by different edge computing devices is different.

[0021] In a sixth aspect, an embodiment of the present application provides an image acquisition system in a drop-and-hook transport scenario, the image acquisition system comprising: a bayonet camera, a fisheye camera, and an electric police camera, wherein different cameras respectively capture different areas of the intersection; The bayonet camera is located at an installation position corresponding to the capture area of ​​the intersection, and collects the front license plate information of the drop-and-hook transport vehicle and uploads it to the corresponding edge computing device; The fisheye camera is located at the installation position corresponding to the middle area of ​​the intersection, and collects the rear license plate information of the drop-and-hook transport vehicle and uploads it to the corresponding edge computing device; The electric police camera is located at an installation position corresponding to the intersection event area, collects image data associated with the drop-and-hook transport vehicle and related events, and uploads the image data to the target edge computing device; There is an overlapping area between the captured area and the middle area, and there is an overlapping area between the event area and the middle area; When the target edge computing device determines that the drop-and-hook transport vehicle triggers a related event based on the image data, the target edge computing device sends an alarm message to the server, so that the server determines the joint identification information of the drop-and-hook transport vehicle based on the attribute information identifier carried in the alarm message, and based on the joint identification information, queries the attribute information of the drop-and-hook transport vehicle from other edge computing devices other than the target edge computing device, wherein the joint identification information includes attribute information identifiers set by different edge computing devices for the drop-and-hook transport vehicle, and the attribute information of the drop-and-hook transport vehicle recorded by different edge computing devices is different.

[0022] In a possible implementation, the image acquisition system further includes: a roadside camera; The roadside camera is located on the side of the intersection, collects side information of the detachable transport vehicle and uploads it to the corresponding edge computing device. The shooting area of ​​the roadside camera overlaps with the middle area.

[0023] In a possible implementation, the driving direction of the drop-and-hook transport vehicle is from the capture area to the middle area or the shooting area of ​​the roadside camera, and from the middle area or the shooting area of ​​the roadside camera to the event area; The order in which different cameras obtain information about the drop-and-hook transport vehicle is the front license plate information collected by the bayonet camera, the rear license plate information collected by the fisheye camera or the side information collected by the roadside camera, and the image data associated with the relevant events collected by the electric police camera.

[0024] In a seventh aspect, an embodiment of the present application provides an electronic device, the electronic device comprising: At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor executes the method in the first aspect or the second aspect above.

[0025] In an eighth aspect, an embodiment of the present application provides a computer storage medium, wherein the computer storage medium stores a computer program, and the computer program is used to enable a computer to execute the method in the first aspect or the second aspect above.

[0026] In a ninth aspect, an embodiment of the present application provides a computer program product, comprising a computer program, which implements the method in the first aspect or the second aspect when the computer program is executed by a processor. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 A schematic diagram of a trailer according to an exemplary embodiment of the present invention; Figure 2A schematic diagram of a trailer according to an exemplary embodiment of the present invention; Figure 3 A schematic diagram of an application scenario of a method for determining vehicle attribute information according to an exemplary embodiment of the present invention; Figure 4 A schematic flow chart of a method for determining vehicle attribute information according to an exemplary embodiment of the present invention; Figure 5 A schematic diagram of a shooting area according to an exemplary embodiment of the present invention; Figure 6 A schematic diagram of another shooting area according to an exemplary embodiment of the present invention; Figure 7 A schematic flow chart of another method for determining vehicle attribute information according to an exemplary embodiment of the present invention; Figure 8 A bird's-eye view according to an exemplary embodiment of the present invention; Fig. 9 A functional schematic diagram of an edge computing device according to an exemplary embodiment of the present invention; Fig.10 A schematic diagram of a device for determining vehicle attribute information according to an exemplary embodiment of the present invention; Fig.11 A schematic diagram of another device for determining vehicle attribute information according to an exemplary embodiment of the present invention; Fig.12 The figure is a schematic diagram of an electronic device according to an exemplary embodiment of the present invention. DETAILED DESCRIPTION

[0028] The principles and spirit of the present application will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided only to enable those skilled in the art to better understand and implement the present application, and are not intended to limit the scope of the present application in any way. On the contrary, these embodiments are provided to make the present disclosure more thorough and complete, and to fully convey the scope of the present disclosure to those skilled in the art.

[0029] Those skilled in the art know that the embodiments of the present application can be implemented as a system, device, method or computer program product. Therefore, the present disclosure can be specifically implemented in the following forms, namely: complete hardware, complete software (including firmware, resident software, microcode, etc.), or a combination of hardware and software.

[0030] It should be understood herein that any number of elements in the drawings is for illustration rather than limitation, and any naming is only for distinction rather than having any limiting meaning.

[0031] The following is an introduction to some concepts involved in the embodiments of the present application.

[0032] Trailer: Figure 1 As shown, the tank is located on the trailer and the license plate information of the tractor and trailer may be different.

[0033] Trailer: As Figure 2 As shown, the tractor may be a small household vehicle connected to the trailer by a towing rope, and the license plate information of the tractor and the license plate information of the trailer may be different.

[0034] Camera: Generally installed at intersections to take pictures of vehicles to obtain image data containing the vehicles.

[0035] Edge computing device: connected to the front-end camera, analyzes the image data collected by the camera, and obtains the attribute information of each vehicle, such as license plate information, number of passengers, vehicle type, etc. The embodiment of the present application does not specifically limit the type of attribute information.

[0036] Target edge computing device: used to determine whether a vehicle has violated traffic regulations based on image data.

[0037] Bird's-eye view: also known as a top view, is obtained by edge computing devices converting image data collected by the camera. The coordinate system of the bird's-eye view obtained by different edge computing devices is the same.

[0038] Related events: Vehicle violation events, such as running a red light, overloading, the target edge computing device has some built-in detection algorithms to determine whether the vehicle triggers a related event. The embodiment of the present application does not specifically limit the type of related events.

[0039] A method for determining vehicle attribute information in an embodiment of the present application is described in detail below with reference to the accompanying drawings.

[0040] like Figure 3 The schematic diagram of an application scenario provided by an embodiment of the present application is shown, and the application scenario includes a server 301, at least one edge computing device (such as Figure 3 The edge computing devices 302_1, 302_2, ..., 302_N shown in the figure are connected to the cameras (such as Figure 3 The server 301 is used to determine the joint identification information of the vehicle, so as to query the attribute information of the vehicle according to the joint identification information of the vehicle; the camera is used to collect the image data of the vehicle; and the edge computing device is used to obtain the attribute information of the vehicle based on the analysis of the acquired image data.

[0041] The server 301 can be an independent physical server or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms. The embodiments of this application do not specifically limit this. The edge computing device of the camera is a device that deploys artificial intelligence algorithms in the user's local area network environment to process video streams from the camera and perform real-time analysis. These devices usually have high computing power and algorithm processing capabilities, and can more effectively handle complex tasks such as license plate recognition and target recognition compared to processing directly on the front-end device. Edge computing devices can be deployed on devices such as edge boxes, alarm hosts, video boxes or gateways. They are responsible for processing video streams and quickly generating results locally, thereby reducing data transmission delays and improving the real-time response capabilities of the system; the embodiments of this application do not limit the types of edge computing devices and cameras.

[0042] In view of the problem that comprehensive vehicle information cannot be obtained at present, the embodiment of the present application provides a method for determining vehicle attribute information, which is applied to the server, such as Figure 4 As shown, the method includes: S401: Receive warning information sent by the target edge computing device when determining that the vehicle triggers a related event.

[0043] The target edge computing device can be connected to the electric police camera, which deploys target detection, tracking algorithms, and event detection algorithms. It can track vehicles based on the images taken by the electric police camera and determine whether related events are triggered. When a vehicle triggers a related event, the target edge computing device will upload an alarm message to the server, indicating that it may be necessary to obtain the attribute information of the vehicle that triggered the related event to clarify the vehicle's status, so that relevant personnel can handle subsequent matters.

[0044] For example, when the target edge computing device determines that a vehicle has run a red light based on the image data captured by the electric police camera, it sends an alarm message to the server to remind relevant personnel to handle the matter.

[0045] S402: Determine the joint identification information of the vehicle based on the attribute information identifier carried in the warning information.

[0046] The above-mentioned alarm information may include the attribute information identification of the vehicle that triggered the relevant event, so that the server can determine the joint identification information of the vehicle based on the attribute information identification of the vehicle, and then obtain more comprehensive attribute information from other edge computing devices based on the joint identification information; the above-mentioned alarm information may also include related events triggered by the vehicle, such as red light running events, overloading events, etc., to facilitate subsequent processing by relevant personnel.

[0047] The above-mentioned joint identification information includes attribute information identifiers set by different edge computing devices for the vehicle. The edge computing device will set an attribute information identifier for each identified vehicle. For example, for any vehicle, the attribute information identifier set by the first edge computing device is "1", the attribute information identifier set by the second edge computing device is "7", and the attribute information identifier set by the third edge computing device is "5". The joint identification information of any vehicle can be "1-7-5". The embodiment of the present application does not limit the specific form of the attribute information identifier and the joint identification information.

[0048] From the above, it can be seen that for any vehicle, the joint identification information is constructed based on the attribute information identification in different edge computing devices. Therefore, it is necessary to determine the attribute information identification set by each edge computing device for any vehicle. In this regard, an embodiment of the present application provides a method for constructing joint identification information based on a bird's-eye view, which specifically includes the following implementation methods.

[0049] For any two edge computing devices, receiving a first bird's-eye view including a first vehicle from one of the any two edge computing devices, and receiving a second bird's-eye view including a second vehicle from another of the any two edge computing devices, wherein the vehicle travels from a first shooting area of ​​a first camera corresponding to the one edge computing device to a second shooting area of ​​a second camera corresponding to the other edge computing device; When the first vehicle and the second vehicle are determined to be the same vehicle based on the generation time of the first bird's-eye view, the coordinates of the first vehicle in the first bird's-eye view, the generation time of the second bird's-eye view, and the coordinates of the second vehicle in the second bird's-eye view, the joint identification information of the vehicles is constructed using the attribute information identifier of the first vehicle obtained from the one edge computing device and the attribute information identifier of the second vehicle obtained from the other edge computing device, wherein the coordinate systems of the bird's-eye views corresponding to different edge computing devices are the same.

[0050] The above bird's-eye view is obtained by converting the image taken by the camera corresponding to the edge computing device. Specifically, the inverse perspective mapping (IPM) method can be used to transform the front view of the image into a top view. The core of the transformation is to convert the points in the image into a unified coordinate system through inverse perspective transformation. The terminal completes the camera calibration. This transformation involves the internal and external parameters of the camera, including the focal length of the camera, the coordinates of the principal point, etc. Based on the detection and tracking performed in the front view or perspective view, the surrounding scene can be represented in the bird's-eye view, which is relatively more intuitive, and the scene understanding and target tracking are more accurate when the target is represented in the bird's-eye view.

[0051] Among them, different cameras have different shooting areas, and the shooting areas corresponding to different cameras may overlap. Figure 5 The image acquisition system in a drop-and-hook transport scenario provided by an embodiment of the present application is shown. Three cameras (which may be a bayonet camera, a fisheye camera, and an electric police camera) are set at the intersection. Different cameras correspond to different shooting areas. The first shooting area (circular) overlaps with the second shooting area (trapezoidal) and the third shooting area (trapezoidal). The driving direction of the vehicle is the second shooting area - the first shooting area - the third shooting area.

[0052] Figure 5 In the three areas where vehicles pass through, bayonet cameras are installed to complete vehicle capture, license plate detection and vehicle tracking in the capture area; fisheye cameras are installed to detect and track vehicles (license plates) after panoramic image correction; electric police cameras are installed in the intersection event area to complete vehicle detection, tracking and event capture. Among them, there is an overlapping area between the capture area and the middle area, and there is an overlapping area between the event area and the middle area.

[0053] In another embodiment, if Figure 6 In the image acquisition system shown, four cameras are set at the intersection (which can be replaced by a bayonet camera, a fisheye camera, an electric police camera, and a roadside camera). Different cameras correspond to different shooting areas, namely, the first shooting area, the second shooting area, the third shooting area, and the fourth shooting area. There are overlapping areas between the shooting areas, namely, the first shooting area overlaps with the second shooting area, the third shooting area overlaps with the second shooting area, and the fourth shooting area overlaps with the second shooting area. That is, there is an overlapping area between the capture area and the middle area, there is an overlapping area between the event area and the middle area, and the shooting area of ​​the roadside camera overlaps with the middle area.

[0054] Figure 6In the 4 areas where the vehicle passes, the bayonet camera can be installed respectively to complete the vehicle capture, license plate detection and vehicle tracking in the capture area; the fisheye camera is installed in the middle area, and the panoramic image is corrected to detect the vehicle (license plate) and track the vehicle; the roadside camera is installed in the middle area to complete the detection of the vehicle side and target capture; the electric police camera is installed in the intersection event area to complete vehicle detection, tracking and event capture. The roadside camera overlaps with the fisheye camera area and only covers the corresponding driving lane. A horizontal capture tripwire is set at the center of the camera. When the center point of the vehicle passes through the tripwire, if it is a trailer or tow truck, the target is captured, and only one capture is taken for the same target; through the fourth camera, the target driving trajectory chain is improved, while the vehicle body image features and structured information are retained, and the integrity of intelligent transportation and intelligent public security information in the smart city is improved. The embodiment of this application does not specifically limit the type of installed cameras and the attribute information collected by each camera.

[0055] In the embodiment of the present application, there are differences in the method of determining the joint identification information corresponding to the case where the vehicle passes through the overlapping area and the case where the vehicle does not pass through the overlapping area.

[0056] by Figure 5 For example, when the vehicle drives to the second shooting area, the edge computing device corresponding to the camera in the second shooting area will generate a vehicle attribute information identifier; when the vehicle drives to the first shooting area, the edge computing device corresponding to the camera in the first shooting area will generate a vehicle attribute information identifier; when the vehicle drives to the third shooting area, the edge computing device corresponding to the camera in the third shooting area will generate a vehicle attribute information identifier. In order to determine that these three identifiers belong to the same vehicle, it is necessary to determine whether the vehicle in the second shooting area is the same vehicle as the vehicle in the first shooting area, and whether the vehicle in the first shooting area is the same vehicle as the vehicle in the third shooting area. Specifically, the following two implementation methods may be included.

[0057] Located in the overlapping area.

[0058] When the generation time of the first bird's-eye view matches the generation time of the second bird's-eye view, and the coordinates of the first vehicle in the first bird's-eye view match the coordinates of the second vehicle in the second bird's-eye view, it is determined that the first vehicle and the second vehicle are the same vehicle.

[0059] by Figure 5 For example, it is determined that the second shooting area corresponds to the first edge computing device, the first shooting area corresponds to the second edge computing device, the first edge computing device corresponds to the first bird's-eye view, the first bird's-eye view includes the first vehicle, and the second edge computing device corresponds to the second bird's-eye view, the second bird's-eye view includes the second vehicle. The implementation method for determining whether the vehicle in the second shooting area is the same vehicle as the vehicle in the first shooting area is as follows: The first edge computing device and the second edge computing device will convert the acquired image data into a bird's-eye view (also called a top view) and send it to the server. If the first vehicle is located in the overlapping area of ​​the second shooting area and the first shooting area, then the bird's-eye view sent by the second edge computing device can be checked for a bird's-eye view with the same generation time and vehicle position. In other words, if the first vehicle is located in the overlapping area in the first bird's-eye view, then the second vehicle in the overlapping area in the second bird's-eye view with the same generation time is the same vehicle as the first vehicle (because the two cameras captured the vehicle at the same time). After determining that the first vehicle and the second vehicle are the same vehicle, the attribute information identifiers of the first vehicle and the second vehicle are associated.

[0060] It should be noted that the attribute information identification can be sent by the edge computing device together with the bird's-eye view, or after determining that the first vehicle and the second vehicle are the same vehicle, the attribute information identification of the first vehicle and the second vehicle can be requested from the edge computing device. This embodiment of the present application does not specifically limit this.

[0061] (2) Not located in the overlapping area.

[0062] In a possible implementation, when the vehicle is driving, it does not pass through the overlapping area, or only passes through the overlapping area for a very small part. In this case, the time when the first vehicle appears in the second shooting area can be predicted based on the generation time of the first bird's-eye view, and the coordinates of the first vehicle when it appears in the second shooting area can be predicted based on the coordinates of the first vehicle in the first bird's-eye view; when the time of appearance matches the generation time of the second bird's-eye view, and the coordinates of the first vehicle when it appears in the second shooting area match the coordinates of the second vehicle in the second bird's-eye view, the first vehicle and the second vehicle are the same vehicle.

[0063] Since the camera corresponding to the second shooting area and the camera corresponding to the first shooting area did not capture the vehicle at the same time, it is necessary to predict the time and position of the vehicle appearing in the second shooting area based on the generation time of the first bird's-eye view and the coordinates of the first vehicle in the first bird's-eye view, and then determine the matching vehicle from the second bird's-eye view, and determine that the vehicles in the two bird's-eye views are the same vehicle, and then associate the attribute information identifiers.

[0064] Since the bird's-eye view is generated in real time, the time when the camera captures the vehicle is not much different from the time when the bird's-eye view is generated, and can be considered to be the same. That is to say, the generation time of the first bird's-eye view can be used to predict the time and coordinates of the first vehicle entering the next shooting area. The prediction process can instruct the edge computing device to perform through the built-in tracking algorithm.

[0065] When constructing joint identification information, if a vehicle passes through N shooting areas, it is necessary to associate N attribute information identifiers. Since the shooting areas passed by the vehicle during driving are in order (for example, the driving direction of the drop-and-hook transport vehicle is from the capture area to the middle area or the shooting area of ​​the roadside camera, and from the middle area or the shooting area of ​​the roadside camera to the event area), that is, the image acquisition data is also in order (the order is: the front license plate information collected by the bayonet camera, the rear license plate information collected by the fisheye camera or the side information collected by the roadside camera, and the image data associated with the relevant event collected by the electric police camera). Therefore, the identifiers corresponding to the two adjacent areas can be associated separately, and then a complete joint identification information can be formed. Figure 5 As shown, the vehicle's driving direction is the second shooting area - the first shooting area - the third shooting area. First, the attribute information identifiers of the second shooting area and the first shooting area are associated to obtain "1-7", and then the attribute information identifiers of the first shooting area and the third shooting area are associated to obtain "7-3", and finally the joint identification information is obtained to obtain "1-7-3". The embodiment of the present application does not limit the order of constructing the joint identification information.

[0066] For example, this method can associate the same vehicle in the overlapping area of ​​2 cameras. First, for each camera perspective, the current position is predicted by the tracking algorithm Mann filter, the prediction box is obtained, and the similarity is calculated. Then, the data is associated by the Hungarian algorithm to assign attribute information identifiers to each vehicle. Finally, by projecting each target in the top view coordinate system, only the vehicle in the overlapping area needs to be associated, and the attribute information identifiers corresponding to the two different cameras of the same vehicle can be obtained, and a new identifier can be fused.

[0067] The bayonet camera is associated with the fisheye camera target. In the last frame of the overlapping area in the bird's-eye view coordinate system, based on the vehicle photographed by the bayonet camera, the vehicles photographed by the fisheye camera are traversed, and the same vehicle attribute information identifier is associated through the maximum intersection ratio and the minimum distance to the target center point. After the association is successful, the fused new identifier (that is, the joint identifier information) is obtained. The relationship between the fused new identifier and the attribute information identifier of the vehicle corresponding to the fisheye camera and the attribute information identifier of the vehicle corresponding to the bayonet camera is maintained.

[0068] The fisheye camera is associated with the electric police camera target. In the last frame of the overlapping area in the bird's-eye view coordinate system, based on the vehicle photographed by the fisheye camera, the vehicles photographed by the electric police camera are traversed, and the same vehicle attribute information identifier is associated through the maximum intersection ratio and the minimum distance to the target center point. After the association is successful, a new fused identifier is obtained, and the relationship between the new identifier and the attribute information identifier of the vehicle corresponding to the fisheye camera and the attribute information identifier of the vehicle corresponding to the electric police camera is maintained to complete the tracing of the same target by multiple cameras.

[0069] For the scenario with four cameras, the edge computing device corresponding to the roadside camera further performs bird's-eye view coordinate conversion and spatial association on the vehicle. The lower edge coordinates of the vehicle's side are spatially associated with the coordinate information of the vehicle captured by the fisheye camera. The fusion identification of the roadside camera and the fisheye camera is obtained through lane position similarity matching.

[0070] The attribute information identification of the same vehicle is associated through the bird's-eye view. Any vehicle captured by the electric police camera can be associated with the same vehicle in the fisheye camera and the bayonet camera through the fusion identification. For related events, it is necessary to provide the target trajectory. In the past, the pictures retained by the single camera evidence were pictures that triggered the related event process. The embodiment of the present application proposes a new evidence picture retention scheme for the related events triggered by trailers and trailers, which retains the pictures and capture times of the related event processes, including 1 picture captured by the electric police camera, 1 picture captured by the fisheye camera, and 1 picture captured by the bayonet camera. The first picture is a picture of the vehicle captured by the bayonet camera; the second picture is a picture of the vehicle appearing in the middle area of ​​the fisheye camera; the third picture is a picture of the vehicle triggering the event when the electric police camera captured the event. At the same time, the edge computing device reports the attribute information identification of the vehicle that triggered the event to the server, and obtains the joint identification information of the vehicle from the attribute information identification, so that all the attribute information of the vehicle can be queried, thereby completing the event penalty of the towing vehicle and the evidence chain of vehicle driving.

[0071] S403: Based on the joint identification information, query the attribute information of the vehicle from other edge computing devices except the target edge computing device.

[0072] Since the edge computing device is connected to the camera, different cameras are installed in different positions, the image data of the vehicle collected is different, and the attribute information of the vehicle obtained is also different. Figure 5 For example, three cameras are installed at the intersection, among which the bayonet camera collects the front license plate information of the vehicle; the fisheye camera collects the front license plate information of the vehicle; the electric police camera captures the vehicle triggering event situation, and the fisheye camera can also collect the rear license plate information of the vehicle, and the bayonet camera can collect the front license plate information. The specific location is determined based on the installation position of the camera, and the embodiment of this application does not specifically limit it. Furthermore, cameras can also be set on the roadside to collect information on the side of the vehicle, such as Figure 6 In addition to the above-mentioned drop-and-hook transport scenario, the image acquisition system provided in the embodiment of the present application can also be applied to other transport scenarios, which are not specifically limited in the embodiment of the present application.

[0073] Afterwards, different cameras will send the collected information to the corresponding edge computing device for processing and recording to obtain the vehicle's attribute information and corresponding identification.

[0074] In one possible implementation, after obtaining the joint identification information, all attribute information of the vehicle can be queried from all edge computing devices, or only part of the attribute information can be queried as needed, which is not specifically limited in this embodiment of the present application.

[0075] Based on the characteristic that the bird's-eye view coordinate systems corresponding to different cameras are the same, the embodiment of the present application associates the attribute information identifiers of the same vehicle in different edge computing devices to obtain the joint identification information of the vehicle, so that relevant personnel can query the attribute information of the vehicle from each edge computing device through the server based on the joint identification information to generate complete evidence.

[0076] Based on the same inventive concept, the embodiment of the present application also provides a method for determining vehicle attribute information, which is applied to edge computing devices, such as Figure 7 As shown, the method includes: S701: Acquire image data of the vehicle through a camera.

[0077] Different cameras have different shooting areas and collect different vehicle attribute information. For example, the bayonet camera in the capture area completes vehicle capture, license plate detection and vehicle tracking; the middle area detects vehicles (license plates) and tracks vehicles; the electric police camera in the intersection event area completes vehicle tracking and event capture; the roadside camera collects vehicle side information. The more cameras deployed at the intersection, the more comprehensive the vehicle attribute information obtained. The embodiment of this application does not specifically limit the number of cameras.

[0078] Each camera corresponds to an edge computing device. After each camera captures an image, it is uploaded to the corresponding edge computing device. Various algorithms are deployed in the edge computing device, such as target detection algorithm, tracking algorithm, driving trajectory analysis algorithm, license plate recognition algorithm, related event recognition, etc., which are not limited in the embodiments of this application.

[0079] S702: When it is determined based on the image data that the vehicle triggers a related event, an alarm message is sent to a server.

[0080] Based on S701, the electric police camera will capture the event, that is, the edge computing device corresponding to the electric police camera records the event triggered by the vehicle. When a vehicle triggers an event, it is necessary to send an alarm message to the server (the alarm message contains the attribute information identifier of the vehicle) so that the server can determine the joint identification information of the vehicle based on the attribute information identifier carried in the alarm message, and then use the joint identification information to obtain comprehensive attribute information about the vehicle to provide complete data for relevant personnel. The implementation method of constructing the joint identification information is as described in S402 above, and will not be repeated here.

[0081] The above-mentioned attribute information identification is generated after the vehicle enters the shooting area of ​​the camera and the corresponding edge computing device identifies the vehicle. In other words, if the vehicle cannot be identified, the attribute information identification cannot be generated for the vehicle. Specifically, the image data collected by the camera can be identified using a pre-built recognition model to determine the type of each vehicle in the image, and each identified vehicle can be identified (different vehicles of the same type have different identifications). The recognition model is trained based on the sample image corresponding to the target type vehicle and the vehicle type corresponding to the sample image.

[0082] In the embodiment of the present application, the main function of the recognition model is to determine which vehicles are included in the image. Therefore, in addition to identifying the type of the above-mentioned vehicle through the recognition model, the recognition model can also be used to generate a bounding box for the detected vehicle to clarify the position of the vehicle in the image. The embodiment of the present application does not limit the specific role of the recognition model, as long as it ensures that each vehicle in the image is marked. In addition, if the recognition model identifies the location of the vehicle, it can also be identified through a bird's-eye view. Figure 8 As shown, the gray rectangle represents a vehicle, and the recognition model identifies the vehicle at position 1 and the vehicle at position 2, and generates attribute information identifiers for each. Since the bird's-eye view is obtained based on the image conversion captured by the camera, after the attribute information identifier is obtained using the bird's-eye view, it can be associated with the original image captured by the camera, and then detailed attribute information can be obtained. The embodiment of the present application does not specifically limit the timing of generating the attribute information identifier.

[0083] The above recognition model can be a target detection model built based on YOLO. Figure 1 The trailer shown and Figure 2 The trailer shown in the figure expands the output detection head of the model and adds two types of training samples. That is to say, in addition to being able to identify ordinary small vehicles, the recognition model can also identify trailers and tow trucks; recognition models can also be constructed separately for trailers and tow trucks, which is not specifically limited in the embodiments of the present application.

[0084] Before adding classification, there would be occasional problems where the vehicle inspection frame only detected the tractor in front, or there would be two detection frames of different sizes for special vehicles, which would lead to unstable tracking and affect the accuracy of event capture. Through the new training method, the stability of the vehicle inspection frame is improved, and the accuracy of event capture is also improved. During the training process, the data is enhanced to the trailer or trailer training set, and on-site feedback pictures are accumulated and simulated. Different tractor types, different carrier types, and different large trailer types are constructed to complete the augmented training set and merge it into the training set of ordinary vehicles. The above target detection model is applicable to image data collected by different cameras.

[0085] In a possible implementation, the edge computing device also performs coordinate system conversion on the image data collected by the camera in real time to obtain a corresponding bird's-eye view. Fig. 9 As shown in the figure, each edge computing device receives the image data sent by its own camera, performs coordinate conversion, and then performs view trajectory fusion, records captured images, and performs task management. Among them, task management can be for event reporting or querying. Through joint identification information, the vehicle capture image, license plate information and time captured by the vehicle bayonet camera, the trajectory target image and time collected by the fisheye camera, and the vehicle trigger event image and time collected by the electric police camera can be traced to form a complete evidence trajectory chain.

[0086] S703: Receive the joint identification information sent by the server, and send the attribute information of the vehicle corresponding to the joint identification information to the server.

[0087] The attribute information of the vehicle is stored in the edge computing device, and the attribute information of the vehicle in different edge computing devices is different. For example, the first edge computing device records the rear license plate information taken by the camera used to collect the rear license plate information; the second edge computing device records the front license plate information taken by the camera used to collect the front license plate information; the third edge computing device records the side information taken by the camera used to collect the side information; the fourth edge computing device records the vehicle that triggers the relevant event taken by the camera used to collect the vehicle triggering relevant events. The embodiments of the present application do not specifically limit the number of edge computing devices and the type of stored attribute information.

[0088] After receiving the joint identification information sent by the server, the corresponding attribute information is sent to the server. For example, if the joint identification information is "1-7-3", the edge computing device corresponds to the value in the middle, that is, 7, and the attribute information with the attribute information identifier of 7 is sent to the server. The joint identification information can also include the identifier of the edge computing device. When the edge computing device recognizes its own identifier, it uses the attribute information identifier corresponding to its own identifier to query the attribute information of the vehicle.

[0089] Based on the same inventive concept, the embodiment of the present application provides an image acquisition system in a drop-and-hook transport scenario, the image acquisition system comprising: a bayonet camera, a fisheye camera and an electric police camera, and different cameras respectively photograph different areas of the intersection; The bayonet camera is located at an installation position corresponding to the capture area of ​​the intersection, and collects the front license plate information of the drop-and-hook transport vehicle and uploads it to the corresponding edge computing device; The fisheye camera is located at the installation position corresponding to the middle area of ​​the intersection, and collects the rear license plate information of the drop-and-hook transport vehicle and uploads it to the corresponding edge computing device; The electric police camera is located at an installation position corresponding to the intersection event area, collects image data associated with the drop-and-hook transport vehicle and related events, and uploads the image data to the target edge computing device; There is an overlapping area between the captured area and the middle area, and there is an overlapping area between the event area and the middle area; When the target edge computing device determines that the detachable and hooked transport vehicle triggers a related event based on the image data, it sends an alarm message to the server, so that the server determines the joint identification information of the detachable and hooked transport vehicle based on the attribute information identifier carried in the alarm message, and based on the joint identification information, queries the attribute information of the detachable and hooked transport vehicle from other edge computing devices other than the target edge computing device, wherein the joint identification information includes attribute information identifiers set by different edge computing devices for the detachable and hooked transport vehicle, and the attribute information of the detachable and hooked transport vehicle recorded by different edge computing devices is different. In an embodiment of the present application, the target edge computing device represents the edge computing device corresponding to the electric police camera.

[0090] The above capture area is as follows Figure 5 The second shooting area in the figure is the camera on the far left, and the middle area is the camera on the far left. Figure 5 The first shooting area in the image is the camera in the middle, and the event area is Figure 5 In the third shooting area, the camera on the far right is the electric police camera. Figure 5 This is just an example provided in the embodiment of the present application. If the camera located in the middle is used to shoot the second shooting area, then the fisheye camera is installed in the middle position. In addition, the fisheye camera can also shoot the second shooting area. The embodiment of the present application does not specifically limit the camera type and the installation position of each camera.

[0091] In a possible implementation, the image acquisition system further includes: a roadside camera; The roadside camera is located on the side of the intersection, collects side information of the detachable transport vehicle and uploads it to the corresponding edge computing device. The shooting area of ​​the roadside camera overlaps with the middle area.

[0092] Roadside cameras are installed in Figure 6 At the position of “1” shown, the shooting area is the third shooting area, and the third shooting area overlaps with the middle area (ie, the second shooting area) shot by the fisheye camera.

[0093] In a possible implementation, the driving direction of the drop-and-hook transport vehicle is from the capture area to the middle area or the shooting area of ​​the roadside camera, and from the middle area or the shooting area of ​​the roadside camera to the event area; The order in which different cameras obtain information about the drop-and-hook transport vehicle is the front license plate information collected by the bayonet camera, the rear license plate information collected by the fisheye camera or the side information collected by the roadside camera, and the image data collected by the electric police camera when the relevant event is triggered.

[0094] by Figure 6 For example, the vehicle travels from bottom to top, and the area it passes through is the first shooting area (checkpoint area) - the second shooting area (middle area) / the third shooting area (roadside camera shooting area) - the fourth shooting area (event area). In other words, the order of acquiring image data is image data taken by the checkpoint camera - image data taken by the fisheye camera / roadside camera - image data taken by the electric police camera.

[0095] The method for determining vehicle attribute information provided in the embodiment of the present application provides a variety of camera installation solutions, which can obtain complete vehicle evidence chain information, and also associates different attribute information identifiers of the same vehicle through the position and size of the vehicle to obtain comprehensive and accurate attribute information of the vehicle. Moreover, this association method does not rely on complete feature structured information and is less affected by environmental interference. At the same time, a recognition model is provided for trailers and trailers, which can identify trailers and trailers.

[0096] Based on the same inventive concept, the embodiment of the present application also provides a method for determining vehicle attribute information, which is applied to the server, such as Fig.10 As shown, the device comprises: The receiving module 1001 is used to receive the warning information sent by the target edge computing device when determining that the vehicle triggers a related event; an identification determination module 1002, configured to determine joint identification information of the vehicle based on the attribute information identifier carried in the warning information, wherein the joint identification information includes attribute information identifiers set by different edge computing devices for the vehicle, the attribute information of the vehicle recorded by the different edge computing devices is different, and the attribute information recorded in the edge computing device is obtained through a camera; The query module 1003 is used to query the attribute information of the vehicle from other edge computing devices except the target edge computing device based on the joint identification information.

[0097] In a possible implementation, the identification module 1002 is used to: For any two edge computing devices, receiving a first bird's-eye view including a first vehicle from one of the any two edge computing devices, and receiving a second bird's-eye view including a second vehicle from another edge computing device of the any two edge computing devices, wherein the bird's-eye view is obtained by converting images captured by cameras corresponding to the edge computing devices, and the vehicle travels from a first capturing area of ​​a first camera corresponding to the one edge computing device to a second capturing area of ​​a second camera corresponding to the other edge computing device; When the first vehicle and the second vehicle are determined to be the same vehicle based on the generation time of the first bird's-eye view, the coordinates of the first vehicle in the first bird's-eye view, the generation time of the second bird's-eye view, and the coordinates of the second vehicle in the second bird's-eye view, the joint identification information of the vehicles is constructed using the attribute information identifier of the first vehicle obtained from the one edge computing device and the attribute information identifier of the second vehicle obtained from the other edge computing device, wherein the coordinate systems of the bird's-eye views corresponding to different edge computing devices are the same.

[0098] In a possible implementation, the identification module 1002 is used to: When the generation time of the first bird's-eye view matches the generation time of the second bird's-eye view, and the coordinates of the first vehicle in the first bird's-eye view match the coordinates of the second vehicle in the second bird's-eye view, it is determined that the first vehicle and the second vehicle are the same vehicle.

[0099] In a possible implementation, the identification module 1002 is used to: Predicting the time when the first vehicle appears in the second shooting area based on the generation time of the first bird's-eye view, and predicting the coordinates of the first vehicle when it appears in the second shooting area based on the coordinates of the first vehicle in the first bird's-eye view; When the time of the appearance matches the generation time of the second bird's-eye view, and the coordinates of the first vehicle when it appears in the second shooting area match the coordinates of the second vehicle in the second bird's-eye view, the first vehicle and the second vehicle are the same vehicle.

[0100] In a possible implementation manner, the query module 1003 is used to determine the attribute information of the vehicle including at least one of the following: The rear license plate information of the vehicle is obtained from the edge computing device corresponding to the camera for collecting the rear license plate information; The front license plate information of the vehicle is obtained from the edge computing device corresponding to the camera used to collect the front license plate information; The side information of the vehicle is obtained from the edge computing device corresponding to the camera used to collect side information.

[0101] Based on the same inventive concept, the embodiment of the present application also provides a method for determining vehicle attribute information, which is applied to edge computing devices, such as Fig.11 As shown, the device comprises: An acquisition module 1101 is used to acquire image data of a vehicle through a camera; A sending module 1102 is configured to send an alarm message to a server when it is determined based on the image data that the vehicle triggers a related event, so that the server determines the joint identification information of the vehicle based on the attribute information identifier carried in the alarm message, wherein the joint identification information includes the attribute information identifier set by different edge computing devices for the vehicle, and the attribute information of the vehicle recorded by different edge computing devices is different; The receiving module 1103 is used to receive the joint identification information sent by the server, and send the attribute information of the vehicle corresponding to the joint identification information to the server through the sending module 1102.

[0102] In a possible implementation, the device further includes an identification module, configured to: Based on the image data, the vehicle is determined to be a target type of vehicle through a pre-constructed recognition model, and an attribute information identifier is generated for the target type of vehicle. The recognition model is trained based on sample images corresponding to the target type of vehicle and the vehicle type corresponding to the sample images.

[0103] Based on the same inventive concept, an embodiment of the present application also provides a system for determining vehicle attribute information, the system comprising: a plurality of cameras, an edge computing device end, and a server end; The camera is used to obtain image data of the vehicle and send the obtained image data to the corresponding edge computing device end, and the camera corresponds to the edge computing device end one by one; The edge computing device is used to receive image data sent by the corresponding camera, and when it is determined based on the image data that the vehicle triggers a related event, send an alarm message to the server; and receive joint identification information sent by the server, and send attribute information of the vehicle corresponding to the joint identification information to the server; The server is used to receive the warning information sent by the target edge computing device when determining that the vehicle triggers a related event; determine the joint identification information of the vehicle based on the attribute information identifier carried in the warning information, and query the attribute information of the vehicle from other edge computing devices except the target edge computing device; The joint identification information includes attribute information identifiers set by different edge computing devices for the vehicle, and the attribute information of the vehicle recorded by different edge computing devices is different.

[0104] Based on the same inventive concept, an embodiment of the present application provides an electronic device, the device comprising: At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute any method for determining vehicle attribute information applied to a server or an edge computing device provided in an embodiment of the present application.

[0105] Refer to the following Fig.12 The electronic device 120 according to this embodiment of the present application is described. Fig.12 The electronic device 120 shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0106] like Fig.12 As shown, the electronic device 120 is in the form of a general electronic device. The components of the electronic device 120 may include but are not limited to: the at least one processor 121, the at least one memory 122, and a bus 123 connecting different system components (including the memory 122 and the processor 121).

[0107] The processor 121 is used to read and execute instructions in the memory 122, so that the at least one processor can execute a method for determining vehicle attribute information provided in the above embodiment and applied to a server or an edge computing device.

[0108] Bus 123 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, a processor, or a local bus using any of a variety of bus architectures.

[0109] The memory 122 may include a readable medium in the form of a volatile memory, such as a random access memory (RAM) 1221 and / or a cache memory 1222 , and may further include a read-only memory (ROM) 1223 .

[0110] The memory 122 may also include a program / utility 1225 having a set (at least one) of program modules 1224, such program modules 1224 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0111] The electronic device 120 may also communicate with one or more external devices 124 (e.g., keyboards, pointing devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 120, and / or communicate with any device that enables the electronic device 120 to communicate with one or more other electronic devices (e.g., routers, modems, etc.). This communication may be performed through an input / output (I / O) interface 125. In addition, the electronic device 120 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) through a network adapter 126. As shown, the network adapter 126 communicates with other modules for the electronic device 120 through a bus 123. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 120, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0112] In some possible implementations, various aspects of a method for determining vehicle attribute information applied to a server or an edge computing device provided by the present application can also be implemented in the form of a program product, which includes program code. When the program product is run on a computer device, the program code is used to enable the computer device to execute the steps of a method for determining vehicle attribute information applied to a server or an edge computing device according to various exemplary embodiments of the present application described above in this specification.

[0113] In addition, the present application also provides a computer-readable storage medium, wherein the computer storage medium stores a computer program, wherein the computer program is used to enable a computer to execute any of the methods described in the above embodiments.

[0114] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0115] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0116] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0117] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A method for determining vehicle attribute information, characterized in that: Applied to the server, the method includes: Receive warning information sent by the target edge computing device when determining that the vehicle triggers a related event; Determine the joint identification information of the vehicle based on the attribute information identifier carried in the warning information, wherein the joint identification information includes attribute information identifiers set by different edge computing devices for the vehicle, the attribute information of the vehicle recorded by the different edge computing devices is different, and the attribute information recorded in the edge computing device is obtained through a camera; Based on the joint identification information, query the attribute information of the vehicle from other edge computing devices except the target edge computing device.

2. The method according to claim 1, characterized in that The vehicle's joint identification information is determined in the following manner: For any two edge computing devices, receiving a first bird's-eye view including a first vehicle from one of the any two edge computing devices, and receiving a second bird's-eye view including a second vehicle from another edge computing device of the any two edge computing devices, wherein the bird's-eye view is obtained by converting images captured by cameras corresponding to the edge computing devices, and the vehicle travels from a first capturing area of ​​a first camera corresponding to the one edge computing device to a second capturing area of ​​a second camera corresponding to the other edge computing device; When the first vehicle and the second vehicle are determined to be the same vehicle based on the generation time of the first bird's-eye view, the coordinates of the first vehicle in the first bird's-eye view, the generation time of the second bird's-eye view, and the coordinates of the second vehicle in the second bird's-eye view, the joint identification information of the vehicles is constructed using the attribute information identifier of the first vehicle obtained from the one edge computing device and the attribute information identifier of the second vehicle obtained from the other edge computing device, wherein the coordinate systems of the bird's-eye views corresponding to different edge computing devices are the same.

3. The method according to claim 2, characterized in that The determining that the first vehicle and the second vehicle are the same vehicle based on the generation time of the first bird's-eye view, the coordinates of the first vehicle in the first bird's-eye view, the generation time of the second bird's-eye view, and the coordinates of the second vehicle in the second bird's-eye view includes: When the generation time of the first bird's-eye view matches the generation time of the second bird's-eye view, and the coordinates of the first vehicle in the first bird's-eye view match the coordinates of the second vehicle in the second bird's-eye view, it is determined that the first vehicle and the second vehicle are the same vehicle.

4. The method according to claim 2, characterized in that: The determining that the first vehicle and the second vehicle are the same vehicle based on the generation time of the first bird's-eye view, the coordinates of the first vehicle in the first bird's-eye view, the generation time of the second bird's-eye view, and the coordinates of the second vehicle in the second bird's-eye view includes: Predicting the time when the first vehicle appears in the second shooting area based on the generation time of the first bird's-eye view, and predicting the coordinates of the first vehicle when it appears in the second shooting area based on the coordinates of the first vehicle in the first bird's-eye view; When the time of the appearance matches the generation time of the second bird's-eye view, and the coordinates of the first vehicle when it appears in the second shooting area match the coordinates of the second vehicle in the second bird's-eye view, the first vehicle and the second vehicle are the same vehicle.

5. The method according to claim 1, characterized in that The vehicle attribute information includes at least one of the following: The rear license plate information of the vehicle is obtained from the edge computing device corresponding to the camera for collecting the rear license plate information; The front license plate information of the vehicle is obtained from the edge computing device corresponding to the camera used to collect the front license plate information; The side information of the vehicle is obtained from the edge computing device corresponding to the camera used to collect side information.

6. A method for determining vehicle attribute information, characterized in that: Applied to edge computing devices, the method includes: Acquire image data of the vehicle through a camera; When it is determined based on the image data that the vehicle triggers a related event, an alarm message is sent to a server, so that the server determines the joint identification information of the vehicle based on the attribute information identifier carried in the alarm message, wherein the joint identification information includes the attribute information identifier set for the vehicle by different edge computing devices, and the attribute information of the vehicle recorded by the different edge computing devices is different; Receive the joint identification information sent by the server, and send the attribute information of the vehicle corresponding to the joint identification information to the server.

7. The method according to claim 6, characterized in that Before determining the vehicle triggering related event based on the image data, the method further includes: Based on the image data, the vehicle is determined to be a target type of vehicle through a pre-constructed recognition model, and an attribute information identifier is generated for the target type of vehicle. The recognition model is trained based on sample images corresponding to the target type of vehicle and the vehicle type corresponding to the sample images.

8. A system for determining vehicle attribute information, characterized in that: The system includes: multiple cameras, an edge computing device end and a server end; The camera is used to obtain image data of the vehicle and send the obtained image data to the corresponding edge computing device end, and the camera corresponds to the edge computing device end one by one; The edge computing device is used to receive image data sent by the corresponding camera, and when it is determined based on the image data that the vehicle triggers a related event, send an alarm message to the server; and receive joint identification information sent by the server, and send attribute information of the vehicle corresponding to the joint identification information to the server; The server is used to receive the warning information sent by the target edge computing device when determining that the vehicle triggers a related event; determine the joint identification information of the vehicle based on the attribute information identifier carried in the warning information, and query the attribute information of the vehicle from other edge computing devices except the target edge computing device; The joint identification information includes attribute information identifiers set by different edge computing devices for the vehicle, and the attribute information of the vehicle recorded by different edge computing devices is different.

9. An image acquisition system in a drop-and-hook transport scenario, characterized in that: The image acquisition system includes: a bayonet camera, a fisheye camera and an electric police camera, and different cameras respectively capture different areas of the intersection; The bayonet camera is located at an installation position corresponding to the capture area of ​​the intersection, and collects the front license plate information of the drop-and-hook transport vehicle and uploads it to the corresponding edge computing device; The fisheye camera is located at the installation position corresponding to the middle area of ​​the intersection, and collects the rear license plate information of the drop-and-hook transport vehicle and uploads it to the corresponding edge computing device; The electric police camera is located at an installation position corresponding to the intersection event area, collects image data associated with the drop-and-hook transport vehicle and related events, and uploads the image data to the target edge computing device; There is an overlapping area between the captured area and the middle area, and there is an overlapping area between the event area and the middle area; When the target edge computing device determines that the drop-and-hook transport vehicle triggers a related event based on the image data, the target edge computing device sends an alarm message to the server, so that the server determines the joint identification information of the drop-and-hook transport vehicle based on the attribute information identifier carried in the alarm message, and based on the joint identification information, queries the attribute information of the drop-and-hook transport vehicle from other edge computing devices other than the target edge computing device, wherein the joint identification information includes attribute information identifiers set by different edge computing devices for the drop-and-hook transport vehicle, and the attribute information of the drop-and-hook transport vehicle recorded by different edge computing devices is different.

10. The image acquisition system according to claim 9, characterized in that: The image acquisition system also includes: a roadside camera; The roadside camera is located on the side of the intersection, collects side information of the detachable transport vehicle and uploads it to the corresponding edge computing device. The shooting area of ​​the roadside camera overlaps with the middle area.

11. The image acquisition system according to claim 9 or 10, characterized in that: The driving direction of the drop-and-hook transport vehicle is from the capture area to the middle area or the shooting area of ​​the roadside camera, and from the middle area or the shooting area of ​​the roadside camera to the event area; The order in which different cameras obtain information about the drop-and-hook transport vehicle is the front license plate information collected by the bayonet camera, the rear license plate information collected by the fisheye camera or the side information collected by the roadside camera, and the image data collected by the electric police camera when the relevant event is triggered.

12. An electronic device, characterized in that: The electronic device comprises: At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor so that the at least one processor executes a method as described in any one of claims 1-5 or 6-7.

13. A computer storage medium, characterized in that: The computer storage medium stores a computer program, and the computer program is used to make a computer execute any one of the methods as claimed in claims 1-5 or 6-7.

14. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 5 or 6 to 7 is implemented.

Images