Vehicle accident handling method, device, computer equipment and storage medium

Abstract

The present application discloses a vehicle accident handling method, device, computer equipment and storage medium, and belongs to the field of artificial intelligence technology. The present application obtains video images of the monitored road section, and determines whether there is a stopped vehicle in the video image. If there is a stopped vehicle, the video clip before the stop time is intercepted, the vehicle trajectory in the video clip is identified, and it is determined whether there is an intersection point in the vehicle trajectory of the stopped vehicle. If there is an intersection point, the stopped vehicle is determined to be an accident vehicle, and driving information is obtained. Based on the vehicle trajectory and driving information, the accident responsibility of the accident vehicle is determined using the traffic control regulations knowledge map, and an accident responsibility document is generated. Finally, evacuation guidance information is generated, and the accident vehicle is instructed to evacuate the accident scene. In addition, the present application also involves blockchain technology, and the owner information can be stored in the blockchain. The present application can significantly improve the efficiency of handling minor vehicle accidents and effectively solve traffic congestion caused by minor vehicle accidents.

Description

Technical Field

[0001] The present application belongs to the field of artificial intelligence technology, and specifically relates to a vehicle accident handling method, device, computer equipment and storage medium. Background Art

[0002] At present, with the rapid development of economy and society, the number of motor vehicles has increased dramatically, and the problem of traffic accidents has become increasingly prominent. Traffic accidents cause casualties and property losses, and also lead to road traffic congestion, and require a large amount of police resources to deal with. According to statistics, in fact, among all traffic accidents, about 80-90% are minor traffic accidents. Minor traffic accidents refer to traffic accidents that occur on the road with motor vehicles, do not involve casualties, and only cause minor property losses. After most minor traffic accidents, the accident vehicle can actually continue to drive.

[0003] According to statistics from relevant departments, when a minor accident occurs, the current approach to handling the accident is still passive, that is, calling the police, receiving the alarm, dispatching the police, handling the accident on the spot, and evacuating. It often requires a large number of police forces to collect evidence and determine responsibility on the spot. The handling process is long, the evacuation is slow, and the handling is slow. The average time from the occurrence to the evacuation is about 40 minutes, which usually causes traffic congestion on the accident section. On urban roads, if the scene is not evacuated within 3 to 5 minutes, it may cause large-scale traffic congestion. How to quickly handle and evacuate after a minor accident has become a practical problem in traffic management. Improving the efficiency of handling minor accidents is imperative. Summary of the invention

[0004] The purpose of the embodiments of the present application is to propose a vehicle accident handling method, device, computer equipment and storage medium to solve the technical problems of existing minor vehicle accidents such as long process, slow evacuation and slow handling, which easily lead to vehicle congestion.

[0005] In order to solve the above technical problems, the embodiment of the present application provides a vehicle accident handling method, which adopts the following technical solutions:

[0006] A vehicle accident handling method, comprising:

[0007] Acquire video images of the monitored road section in real time, and determine whether there is a suddenly stopped vehicle in the video images based on a pre-trained content recognition model;

[0008] If there is a suddenly stopped vehicle in the video image, determining the stopping time of the suddenly stopped vehicle based on the time axis of the video image;

[0009] intercepting a video segment within a predetermined time interval before the stopping moment from the video image, and identifying a vehicle trajectory of the suddenly stopped vehicle from the video segment;

[0010] Determining whether there is an intersection point in the vehicle trajectory of the suddenly stopped vehicle;

[0011] If the vehicle trajectory of the suddenly stopped vehicle has an intersection, then the suddenly stopped vehicle is determined to be an accident vehicle;

[0012] Acquire driving information of the accident vehicle, and determine the accident responsibility of the accident vehicle using a pre-built traffic control regulations knowledge graph based on the vehicle trajectory and the driving information;

[0013] Acquiring the owner information of the accident vehicle, and generating an accident responsibility document based on the owner information and the accident responsibility;

[0014] Evacuation guidance information is generated based on the vehicle trajectory, and the accident vehicle is instructed to evacuate the accident scene according to the evacuation guidance information.

[0015] Furthermore, the step of acquiring video images of the monitored road section in real time and determining whether there is a suddenly stopped vehicle in the video images based on a pre-trained content recognition model specifically includes:

[0016] Cutting the video images according to a preset time interval to obtain a video to be identified;

[0017] Extracting key frames from the video to be identified to obtain key frame images;

[0018] The key frame image is input into the content recognition model to determine whether there is a vehicle that suddenly stops in the video image.

[0019] Furthermore, the step of determining the stopping time of the suddenly stopped vehicle based on the time axis of the video image specifically includes:

[0020] Obtaining frame number information of the key frame image;

[0021] The stopping time of the suddenly stopped vehicle is determined based on the frame number information and the time axis of the video to be identified.

[0022] Furthermore, the step of capturing a video clip within a predetermined time interval before the stopping moment from the video image, and identifying the vehicle trajectory of the suddenly stopped vehicle from the video clip specifically includes:

[0023] Extracting a video clip within a predetermined time interval before the parking moment from the video image;

[0024] Extracting video frames of the video clip;

[0025] The video frames are imported into a pre-trained trajectory recognition model according to the time sequence of the video clips to obtain the vehicle trajectory of the suddenly stopped vehicle.

[0026] Furthermore, the step of importing the video frames into a pre-trained trajectory recognition model according to the time sequence of the video clips to obtain the vehicle trajectory of the suddenly stopped vehicle specifically includes:

[0027] Extracting features from the video frame to obtain a feature sequence;

[0028] Performing vector transformation on the feature sequence to obtain a feature vector sequence;

[0029] splicing the feature vector sequence to obtain an initial feature matrix;

[0030] Performing a convolution operation on the initial feature matrix to obtain a convolution feature matrix;

[0031] The convolution feature matrices are combined to obtain the vehicle trajectory of the suddenly stopped vehicle.

[0032] Furthermore, the step of determining the accident responsibility of the accident vehicle using a pre-built traffic control regulations knowledge graph based on the vehicle trajectory and the driving information specifically includes:

[0033] Constructing an associated entity of the accident vehicle based on the vehicle trajectory and the driving information;

[0034] The associated entity is imported into the traffic control regulations knowledge graph, and the accident responsibility of the accident vehicle is output.

[0035] Furthermore, after the step of obtaining the owner information of the accident vehicle and generating an accident responsibility document based on the owner information and the accident responsibility, it also includes:

[0036] Generating a guidance interface for accident claim settlement operations, and displaying the guidance interface to the owner of the accident vehicle;

[0037] In response to the operation of the owner of the accident vehicle on the guidance interface, generating claim settlement operation steps, wherein the claim settlement operation steps at least include taking photos for evidence collection and damage assessment analysis;

[0038] Generate a framing frame, and instruct the owner of the accident vehicle to use the framing frame to complete photo collection and upload vehicle damage photos according to the claim settlement operation steps on the guidance interface;

[0039] Using a pre-trained vehicle damage recognition model to recognize the vehicle damage photo, and obtaining the degree of vehicle damage;

[0040] A compensation plan is generated based on the degree of vehicle damage and the accident liability.

[0041] In order to solve the above technical problems, the embodiment of the present application also provides a vehicle accident handling device, which adopts the following technical solution:

[0042] A vehicle accident handling device, comprising:

[0043] The vehicle sudden stop monitoring module is used to obtain the video image of the monitored road section in real time and determine whether there is a sudden stop vehicle in the video image based on a pre-trained content recognition model;

[0044] A parking time acquisition module, used for determining the parking time of the suddenly stopped vehicle based on the time axis of the video image when there is a suddenly stopped vehicle in the video image;

[0045] A vehicle track recognition module, used to capture a video clip within a predetermined time interval before the stopping moment from the video image, and recognize the vehicle track of the suddenly stopped vehicle from the video clip;

[0046] A track intersection judgment module is used to judge whether there is an intersection point in the vehicle track of the suddenly stopped vehicle;

[0047] An accident vehicle confirmation module, used for determining that the suddenly stopped vehicle is an accident vehicle if there is an intersection in the vehicle trajectory of the suddenly stopped vehicle;

[0048] An accident responsibility determination module, used to obtain the driving information of the accident vehicle, and determine the accident responsibility of the accident vehicle based on the vehicle trajectory and the driving information using a pre-built traffic control regulations knowledge graph;

[0049] A responsibility document generation module, used to obtain the owner information of the accident vehicle, and generate an accident responsibility document based on the owner information and the accident responsibility;

[0050] The vehicle evacuation guidance module is used to generate evacuation guidance information based on the vehicle trajectory and instruct the accident vehicle to evacuate the accident scene according to the evacuation guidance information.

[0051] In order to solve the above technical problems, the embodiment of the present application further provides a computer device, which adopts the following technical solution:

[0052] A computer device comprises a memory and a processor, wherein the memory stores computer-readable instructions, and the processor implements the steps of any one of the above-mentioned vehicle accident handling methods when executing the computer-readable instructions.

[0053] In order to solve the above technical problems, the embodiment of the present application further provides a computer-readable storage medium, which adopts the following technical solution:

[0054] A computer-readable storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by a processor, the steps of the vehicle accident handling method as described in any one of the above are implemented.

[0055] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0056] The present application discloses a vehicle accident handling method, device, computer equipment and storage medium, which belongs to the field of artificial intelligence technology. The present application obtains video images of the monitored road section in real time, and determines whether there is a stopped vehicle in the video image. If there is a stopped vehicle in the image, the stopping time of the stopped vehicle is immediately determined, and a video clip within a predetermined time interval before the stopping time is intercepted from the video image, and the vehicle trajectory of the stopped vehicle is identified from the video clip, and it is determined whether there is an intersection point in the vehicle trajectory of the stopped vehicle. If there is an intersection point in the vehicle trajectory of the stopped vehicle, the stopped vehicle is determined to be an accident vehicle, the driving information of the accident vehicle is obtained, and the accident responsibility of the accident vehicle is determined based on the vehicle trajectory and driving information using a pre-built traffic control regulations knowledge graph, the owner information of the accident vehicle is obtained, and the accident responsibility document is generated based on the owner information and the accident responsibility, and the evacuation guidance information is generated based on the vehicle trajectory, and the accident vehicle is instructed to evacuate the accident scene according to the evacuation guidance information. This application provides a process for handling minor vehicle accidents. By identifying the stopped vehicles and vehicle trajectories, and combining the knowledge graph of traffic regulations to automatically determine the accident responsibility of the accident vehicle, the efficiency of handling minor vehicle accidents is significantly improved, and traffic congestion caused by minor vehicle accidents is effectively solved. In addition, this application also provides an automatic claims settlement process for minor vehicle accidents, further improving the efficiency of handling minor vehicle accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the scheme in the present application, a brief introduction is given below to the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0058] Figure 1 An exemplary system architecture diagram is shown in which the present application can be applied;

[0059] Figure 2 A flow chart showing an embodiment of a vehicle accident handling method according to the present application;

[0060] Figure 3 A schematic structural diagram of an embodiment of a vehicle accident handling device according to the present application is shown;

[0061] Figure 4 A schematic structural diagram of an embodiment of a computer device according to the present application is shown. DETAILED DESCRIPTION

[0062] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by technicians in the technical field of the present application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit the present application; the terms "including" and "having" and any variations thereof in the specification and claims of the present application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of the present application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0063] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0064] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0065] like Figure 1 As shown, the system architecture 100 may include terminal devices 101, 102, 103, a network 104 and a server 105. The network 104 is used to provide a medium for communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0066] Users can use terminal devices 101, 102, 103 to interact with server 105 through network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, 103, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0067] Terminal devices 101, 102, 103 can be various electronic devices with display screens and supporting web browsing, including but not limited to smart phones, tablet computers, e-book readers, MP3 players (Moving Picture Experts Group Audio Layer III), MP4 (Moving Picture Experts Group Audio Layer IV) players, laptop computers, desktop computers, etc.

[0068] Server 105 can be a server that provides various services, such as a background server that provides support for pages displayed on terminal devices 101, 102, and 103. The server can be an independent server 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 (CDNs), and big data and artificial intelligence platforms.

[0069] It should be noted that the vehicle accident handling method provided in the embodiment of the present application is generally executed by a server, and accordingly, the vehicle accident handling device is generally arranged in the server.

[0070] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0071] Continue to refer Figure 2 , showing a flow chart of an embodiment of the vehicle accident handling method according to the present application. The embodiment of the present application can acquire and process relevant data based on artificial intelligence technology. Among them, artificial intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0072] The basic technologies of artificial intelligence generally include sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics and other technologies. Artificial intelligence software technology mainly includes computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning / deep learning. The vehicle accident handling method includes the following steps:

[0073] S201, acquiring video images of a monitored road section in real time, and determining whether there is a suddenly stopped vehicle in the video images based on a pre-trained content recognition model.

[0074] Specifically, high-definition cameras are installed at important traffic sections (such as intersections with heavy traffic, highway ramps, etc.), and video images of important traffic sections are obtained in real time through high-definition, and uploaded to the server. The server receives the accident processing request, obtains the video images of the monitored section in real time, and determines whether there is a stopped vehicle in the video image based on the pre-trained content recognition model.

[0075] It should be noted that in the specific embodiments of the present application, the content recognition model is a pre-trained machine learning model specifically for vehicle sudden stops. The training data used by the content recognition model is a series of pre-collected videos of vehicle sudden stops, where the video conditions of vehicle sudden stops include videos of sudden stops without collisions and videos of sudden stops with collisions. A content recognition model that can realize vehicle sudden stop recognition is pre-trained through supervised training.

[0076] In this embodiment, the vehicle accident handling method is executed on the electronic device (eg Figure 1 The server shown in the figure) can receive the vehicle accident handling request through a wired connection or a wireless connection. It should be noted that the above wireless connection method may include but is not limited to 3G / 4G connection, WiFi connection, Bluetooth connection, WiMAX connection, Zigbee connection, UWB (ultra wideband) connection, and other wireless connection methods currently known or to be developed in the future.

[0077] S202: If there is a suddenly stopped vehicle in the video image, determine the stopping time of the suddenly stopped vehicle based on the time axis of the video image.

[0078] Specifically, if a suddenly stopped vehicle is identified in the video image, a video frame corresponding to the suddenly stopped screen is determined, and the stopping time of the suddenly stopped vehicle is determined based on the video frame corresponding to the suddenly stopped screen and the time axis of the video image.

[0079] S203, capturing a video clip within a predetermined time interval before the stopping moment from the video image, and identifying a vehicle trajectory of the suddenly stopped vehicle from the video clip.

[0080] Specifically, a video clip within a predetermined time interval before the parking moment is intercepted from the video image, and the vehicle trajectory of the suddenly stopped vehicle is identified from the video clip. The predetermined time interval can be selected according to actual needs, such as setting the predetermined time interval for ordinary roads to 10 minutes and for high-speed roads to 5 minutes. By intercepting the video clip before the parking moment and using the video clip to complete the vehicle trajectory, the amount of data processing can be reduced while ensuring the accuracy of trajectory recognition.

[0081] S204, determining whether there is an intersection point in the vehicle track of the suddenly stopped vehicle.

[0082] Specifically, the server uses the vehicle trajectory analysis results to determine whether there is an intersection in the vehicle trajectories of the suddenly stopped vehicles. If there is an intersection in the vehicle trajectories, it proves that a traffic accident has occurred. If there is no intersection in the vehicle trajectories, it proves that it is just a regular stop. For example, two cars have an accident, and the camera monitors the sudden stop of the vehicle. Through vehicle trajectory analysis, it is found that the vehicle trajectories of the two cars intersect, and it is determined that a traffic accident has occurred. Then, the severity of the accident is determined based on the accident pictures taken by the camera. If it is a major accident, an alarm is automatically initiated. If it is a minor accident, the identity of the accident owner is identified through identity recognition, and a minor accident handling guidance process is provided to the accident owner to improve the efficiency of minor accident handling.

[0083] It should be noted that before determining the vehicle trajectory of a suddenly stopped vehicle, the server first determines the number of suddenly stopped vehicles. For example, if the number of suddenly stopped vehicles is 1, it is determined that it is not a vehicle collision accident, and the suddenly stopped vehicles are continuously monitored. When the parking time of the suddenly stopped vehicle exceeds the preset threshold, an inquiry message is sent to the owner's terminal to ask whether the owner needs assistance.

[0084] S205: If there is an intersection point in the vehicle trajectory of the suddenly stopped vehicle, it is determined that the suddenly stopped vehicle is an accident vehicle.

[0085] Specifically, if there is an intersection in the vehicle trajectory of the suddenly stopped vehicle, the suddenly stopped vehicle is determined to be an accident vehicle, and the on-site camera is further called to take pictures of the accident scene. The extent of the accident is judged based on the pictures of the accident scene. If it is a major accident, an alarm is automatically initiated. If it is a minor accident, the identity of the accident vehicle owner is identified through the person-vehicle associated identity recognition system, and the accident vehicle owner is provided with an accident rule guidance process, an accident claim guidance process, and a minor accident handling guidance process.

[0086] S206, obtaining driving information of the accident vehicle, and determining the accident responsibility of the accident vehicle based on the vehicle trajectory and the driving information using a pre-built traffic control regulations knowledge graph.

[0087] Specifically, when it is determined that the vehicle accident is a minor accident, the driving information of the accident vehicle is obtained from the video clip, such as the vehicle speed before the accident, the lane change situation before the accident, etc., and the accident responsibility of the accident vehicle is determined based on the vehicle trajectory and driving information using a pre-built traffic control regulations knowledge graph.

[0088] S207, obtaining the owner information of the accident vehicle, and generating an accident responsibility document based on the owner information and the accident responsibility.

[0089] Specifically, the server obtains the owner information of the accident vehicle, generates an accident responsibility document based on the owner information and the accident responsibility, and pushes the accident responsibility document to the owner of the accident vehicle for the owner to review.

[0090] S208, generating evacuation guidance information based on the vehicle trajectory, and instructing the accident vehicle to evacuate the accident scene according to the evacuation guidance information.

[0091] Specifically, the server generates evacuation guidance information based on the vehicle trajectory, and instructs the accident vehicle to evacuate the accident scene according to the evacuation guidance information, so as to reduce traffic congestion caused by minor vehicle accidents.

[0092] It should be noted that in order to further reduce traffic congestion caused by minor vehicle accidents, the step of guiding information to evacuate the accident scene can be executed before the steps of accident rules and vehicle damage compensation. That is, after a minor vehicle accident occurs, the server guides the owner of the accident vehicle to complete the report and take photos of the vehicle damage, and immediately generates evacuation guidance information, and instructs the accident vehicle to evacuate the accident scene according to the evacuation guidance information. The process of accident rules and vehicle damage compensation is automatically processed by the server in the background and sent to the owner of the accident vehicle via text message, effectively shortening the processing time of minor vehicle accidents to within 5 minutes, thereby improving the efficiency of handling minor vehicle accidents.

[0093] In the above embodiment, the present application provides a processing flow for minor vehicle accidents. By real-time identification of suddenly stopped vehicles and vehicle trajectories, and combining the knowledge graph of traffic management regulations to automatically determine the accident responsibility of the accident vehicle, the efficiency of handling minor vehicle accidents is significantly improved, and traffic congestion caused by minor vehicle accidents is effectively solved.

[0094] Furthermore, the step of acquiring the video image of the monitored road section in real time and judging whether there is a suddenly stopped vehicle in the video image based on a pre-trained content recognition model specifically includes:

[0095] Cutting the video images according to a preset time interval to obtain a video to be identified;

[0096] Extracting key frames from the video to be identified to obtain key frame images;

[0097] The key frame image is input into the content recognition model to determine whether there is a vehicle that suddenly stops in the video image.

[0098] Specifically, the server cuts the video image according to a preset interval length to segment the video image to obtain a video to be identified, extracts key frames from the video to be identified to obtain a key frame image, and inputs the key frame image into a content recognition model to determine whether there is a vehicle that suddenly stops in the video image.

[0099] The preset time interval can be set to 10s, 30s, 60s, etc. By processing the video in segments, the amount of data processing can be effectively reduced, the consumption of computing resources can be reduced, and the pressure on the server can be reduced. By obtaining the key frame images of the video to be identified, judging whether there is a stopped vehicle in the video image through the key frame images can also effectively reduce the amount of data processing.

[0100] It should be noted that the content recognition model of this embodiment is a pre-trained artificial intelligence model specifically for vehicle sudden stops, and the training data used is a series of pre-collected videos of vehicle sudden stops, where the video conditions of vehicle sudden stops include videos of sudden stops without collisions and videos of sudden stops with collisions. A content recognition model that can realize vehicle sudden stop recognition is trained by using the pre-collected videos of vehicle sudden stops.

[0101] In the above embodiment, the present application cuts the video image and then sequentially identifies the videos to be identified after cutting to determine whether there is a vehicle that suddenly stops in the video image, thereby effectively reducing the amount of data processing and reducing the pressure on the server.

[0102] Furthermore, the step of determining the stopping time of the suddenly stopped vehicle based on the time axis of the video image specifically includes:

[0103] Obtaining frame number information of the key frame image;

[0104] The stopping time of the suddenly stopped vehicle is determined based on the frame number information and the time axis of the video to be identified.

[0105] Specifically, when there is a vehicle that stops suddenly in the video image, the frame number information of the key frame image is obtained, the total number of frames of the video to be identified and the time axis of the video to be identified are determined, and the stopping time of the vehicle that stops suddenly is determined according to the frame number information of the key frame image, the total number of frames of the video to be identified and the time axis of the video to be identified. For example, if the time axis of the video to be identified is from 15:35:01 to 15:35:60, the total number of frames of the video to be identified is 60 frames, and the frame number information of the key frame image shows that the key frame image is the 45th frame, then the stopping time of the vehicle that stops suddenly is determined to be 15:35:45.

[0106] In the above embodiment, the present application calculates the stopping time of the suddenly stopped vehicle by taking the frame number information of the key frame image and the total frame number of the video to be identified, so as to facilitate the subsequent capture of accurate video clips before and after the vehicle accident.

[0107] Furthermore, the step of capturing a video clip within a predetermined time interval before the stopping moment from the video image, and identifying the vehicle trajectory of the suddenly stopped vehicle from the video clip specifically includes:

[0108] Extracting a video clip within a predetermined time interval before the parking moment from the video image;

[0109] Extracting video frames of the video clip;

[0110] The video frames are imported into a pre-trained trajectory recognition model according to the time sequence of the video clips to obtain the vehicle trajectory of the suddenly stopped vehicle.

[0111] Among them, the pre-trained trajectory recognition model is built based on the convolutional neural network structure. Convolutional Neural Networks (CNN) is a type of feedforward neural network with deep structure that includes convolution calculations. It is one of the representative algorithms of deep learning. Convolutional neural networks have representation learning capabilities and can perform shift-invariant classification of input information according to their hierarchical structure. Therefore, they are also called "shift-invariant artificial neural networks." Convolutional neural networks are constructed by imitating the visual perception mechanism of organisms. They can perform supervised learning and unsupervised learning. The convolution kernel parameter sharing within the convolution layer and the sparsity of the inter-layer connection enable convolutional neural networks to learn grid-like topology features such as pixels and audio with a small amount of computation. It has a stable effect and does not require additional feature engineering for the data.

[0112] Specifically, the server captures a video clip within a predetermined time interval before the parking moment from the video image, extracts video frames of the video clip, obtains a video frame sequence collection with time series characteristics, and imports the video frame sequence collection with time series characteristics into a pre-trained trajectory recognition model. The trajectory recognition model recognizes the driving path in the video frame sequence collection to obtain the vehicle trajectory of the suddenly stopped vehicle.

[0113] Furthermore, the step of importing the video frames into a pre-trained trajectory recognition model according to the time sequence of the video clips to obtain the vehicle trajectory of the suddenly stopped vehicle specifically includes:

[0114] Extracting features from the video frame to obtain a feature sequence;

[0115] Performing vector transformation on the feature sequence to obtain a feature vector sequence;

[0116] splicing the feature vector sequence to obtain an initial feature matrix;

[0117] Performing a convolution operation on the initial feature matrix to obtain a convolution feature matrix;

[0118] The convolution feature matrices are combined to obtain the vehicle trajectory of the suddenly stopped vehicle.

[0119] Specifically, the trajectory recognition model includes an input layer, a convolution layer and an output layer, wherein the input layer performs feature extraction on the video frame to obtain a feature sequence, the feature sequence is vectorized to obtain a feature vector sequence, the feature vector sequence is spliced ​​to obtain an initial feature matrix, the initial feature matrix is ​​convolved in the convolution layer to obtain a convolution feature matrix, the local features of the vehicle trajectory are obtained, and the convolution feature matrix is ​​combined in the output layer to obtain the vehicle trajectory of the suddenly stopped vehicle.

[0120] It should be noted that the convolution calculation process of the convolution layer uses an m*n matrix. Taking 1D convolution as an example, a x*n convolution kernel is constructed. The convolution kernel slides on the initial feature vector to obtain multiple local feature matrices. The initial feature vector is a multidimensional vector. For example, if the value of m is 5 and the value of x is 1, the convolution kernel slides from top to bottom. x is first multiplied and summed with the n-dimensional vector in the first row to obtain a value. Then x continues to slide down and convolve with the second row, the third row... to obtain a 5*1 matrix, which is the convolution result. The output layer is a fully connected layer, which is used to integrate the local feature matrix generated by the convolution layer. The local features of the vehicle trajectory in the video frame sequence collection are selected through the convolution layer, and then the local features are reassembled into complete features through the weight matrix through the output layer to obtain the vehicle trajectory of the stopped vehicle.

[0121] In the above embodiment, the present application builds a trajectory recognition model through a convolutional neural network structure, and processes a collection of video frame sequences with time series characteristics through the trajectory recognition model to obtain the vehicle trajectory of the suddenly stopped vehicle.

[0122] Furthermore, the step of determining the accident responsibility of the accident vehicle using a pre-built traffic control regulations knowledge graph based on the vehicle trajectory and the driving information specifically includes:

[0123] Constructing an associated entity of the accident vehicle based on the vehicle trajectory and the driving information;

[0124] The associated entity is imported into the traffic control regulations knowledge graph, and the accident responsibility of the accident vehicle is output.

[0125] Among them, knowledge graph, as a semantic network, has extremely strong expressive power and modeling flexibility. First, knowledge graph is a semantic representation that can model entities, concepts, attributes and the relationships between them in the real world; second, knowledge graph is the data exchange standard of its derivative technology, which is itself a "protocol" for data modeling. Related technologies cover various aspects such as knowledge extraction, knowledge integration, knowledge management and knowledge application.

[0126] Specifically, the server pre-builds a traffic regulations knowledge graph through the collected historical vehicle accident event information, and the traffic regulations knowledge graph is used to quickly determine the responsibility of the vehicle accident. When determining the vehicle accident rules, the server obtains the pre-built traffic regulations knowledge graph, and builds the associated entities of the accident vehicle based on the driving information of the accident vehicle and the vehicle trajectory of the accident vehicle, such as "accident vehicle-violation-speeding", "accident vehicle-violation-illegal change of direction", etc., and determines the accident responsibility of the accident vehicle based on the constructed associated entities of the accident vehicle and the traffic regulations knowledge graph.

[0127] Furthermore, after the step of obtaining the owner information of the accident vehicle and generating an accident responsibility document based on the owner information and the accident responsibility, it also includes:

[0128] Generating a guidance interface for accident claim settlement operations, and displaying the guidance interface to the owner of the accident vehicle;

[0129] In response to the operation of the owner of the accident vehicle on the guidance interface, generating claim settlement operation steps, wherein the claim settlement operation steps at least include taking photos for evidence collection and damage assessment analysis;

[0130] Generate a framing frame, and instruct the owner of the accident vehicle to use the framing frame to complete photo collection and upload vehicle damage photos according to the claim settlement operation steps on the guidance interface;

[0131] Using a pre-trained vehicle damage recognition model to recognize the vehicle damage photo, and obtaining the degree of vehicle damage;

[0132] A compensation plan is generated based on the degree of vehicle damage and the accident liability.

[0133] Specifically, after generating the accident responsibility document, the server can also guide the car owner to automatically complete the vehicle damage claim. First, the server generates a guidance interface for the accident claim operation and displays the guidance interface to the owner of the accident vehicle. Then, in response to the operation of the owner of the accident vehicle on the guidance interface, the claim operation steps are generated, wherein the claim operation steps at least include taking photos for evidence collection and damage assessment analysis. A viewfinder is generated on the car owner's terminal display, and according to the claim operation steps on the guidance interface, the owner of the accident vehicle is instructed to use the viewfinder to complete and upload the vehicle damage photos, and the vehicle damage photos are identified using the pre-trained vehicle damage recognition model to obtain the degree of vehicle damage, and a compensation plan is generated based on the degree of vehicle damage and accident responsibility.

[0134] It should be noted that the vehicle damage identification model is also trained based on the machine learning model. The training data used is a series of pre-collected vehicle damage photos, and the vehicle damage features are pre-marked on the vehicle damage photos. The corresponding compensation plan is matched according to the vehicle damage features, and the vehicle damage photos and the compensation plan corresponding to the vehicle damage photos are imported into the machine learning model to obtain the vehicle damage identification model.

[0135] Furthermore, the vehicle accident handling method further includes:

[0136] Identifying the driving direction of the accident vehicle and determining the opposite direction of the driving direction;

[0137] generating evacuation guidance information of the accident vehicle based on the vehicle trajectory and the opposite direction of the driving direction;

[0138] Based on the evacuation guidance information, the accident vehicle is instructed to evacuate the accident scene in the opposite direction of the driving direction.

[0139] Specifically, the server identifies the driving direction of the accident vehicle and determines the opposite direction of the driving direction, generates evacuation guidance information for the accident vehicle based on the vehicle trajectory and the opposite direction of the driving direction, and instructs the accident vehicle to evacuate the accident scene in the opposite direction of the driving direction based on the evacuation guidance information.

[0140] It should be noted that in order to avoid traffic congestion caused by traffic accidents, the step of guiding the evacuation of accident vehicles can occur before the accident rules and accident claims. After the accident occurs, the owner of the accident vehicle only needs to take detailed pictures of the accident scene according to the guidance. When the pictures are successfully reviewed, the vehicle evacuation procedure can be executed. Accident rules and accident claims can be processed after the accident vehicle is evacuated, further reducing the vehicle congestion caused by the accident.

[0141] In this embodiment, the present application discloses a vehicle accident handling method, which belongs to the field of artificial intelligence technology. The present application obtains video images of the monitored road section in real time, and determines whether there is a stopped vehicle in the video image. If there is a stopped vehicle in the image, the stopping time of the stopped vehicle is immediately determined, and a video clip within a predetermined time interval before the stopping time is intercepted from the video image, and the vehicle trajectory of the stopped vehicle is identified from the video clip, and it is determined whether there is an intersection point in the vehicle trajectory of the stopped vehicle. If there is an intersection point in the vehicle trajectory of the stopped vehicle, the stopped vehicle is determined to be an accident vehicle, and the driving information of the accident vehicle is obtained, and the accident responsibility of the accident vehicle is determined based on the vehicle trajectory and driving information using a pre-built traffic control regulations knowledge graph, the owner information of the accident vehicle is obtained, and the accident responsibility document is generated based on the owner information and the accident responsibility, and the evacuation guidance information is generated based on the vehicle trajectory, and the accident vehicle is instructed to evacuate the accident scene according to the evacuation guidance information. This application provides a process for handling minor vehicle accidents. By identifying the stopped vehicles and vehicle trajectories, and combining the knowledge graph of traffic regulations to automatically determine the accident responsibility of the accident vehicle, the efficiency of handling minor vehicle accidents is significantly improved, and traffic congestion caused by minor vehicle accidents is effectively solved. In addition, this application also provides an automatic claims settlement process for minor vehicle accidents, further improving the efficiency of handling minor vehicle accidents.

[0142] It should be emphasized that in order to further ensure the privacy and security of the above-mentioned car owner information, the above-mentioned car owner information can also be stored in a blockchain node.

[0143] The blockchain referred to in this application is a new application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, encryption algorithm, etc. Blockchain is essentially a decentralized database, a string of data blocks generated by cryptographic methods. Each data block contains a batch of network transaction information, which is used to verify the validity of its information (anti-counterfeiting) and generate the next block. Blockchain can include the underlying blockchain platform, platform product service layer, and application service layer.

[0144] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing related hardware through computer-readable instructions, and the computer-readable instructions can be stored in a computer-readable storage medium. When the computer-readable instructions are executed, they can include the processes of the embodiments of the above-mentioned methods. Among them, the aforementioned storage medium can be a non-volatile storage medium such as a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0145] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0146] Further references Figure 3 , as a response to the above Figure 2 The present application provides an embodiment of a vehicle accident handling device, and the device embodiment is similar to Figure 2 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0147] like Figure 3 As shown, the vehicle accident handling device described in this embodiment includes:

[0148] The vehicle sudden stop monitoring module 301 is used to obtain the video image of the monitored road section in real time, and determine whether there is a sudden stop vehicle in the video image based on a pre-trained content recognition model;

[0149] A parking time acquisition module 302 is used to determine the parking time of the suddenly stopped vehicle based on the time axis of the video image when there is a suddenly stopped vehicle in the video image;

[0150] The vehicle track recognition module 303 is used to capture a video clip within a predetermined time interval before the stopping moment from the video image, and recognize the vehicle track of the suddenly stopped vehicle from the video clip;

[0151] A track intersection determination module 304 is used to determine whether the vehicle track of the suddenly stopped vehicle has an intersection point;

[0152] The accident vehicle confirmation module 305 is used to determine that the suddenly stopped vehicle is an accident vehicle if there is an intersection in the vehicle trajectory of the suddenly stopped vehicle;

[0153] The accident responsibility determination module 306 is used to obtain the driving information of the accident vehicle, and determine the accident responsibility of the accident vehicle based on the vehicle trajectory and the driving information using a pre-built traffic control regulations knowledge graph;

[0154] A responsibility document generation module 307 is used to obtain the owner information of the accident vehicle and generate an accident responsibility document based on the owner information and the accident responsibility;

[0155] The vehicle evacuation guidance module 308 is used to generate evacuation guidance information based on the vehicle trajectory, and instruct the accident vehicle to evacuate the accident scene according to the evacuation guidance information.

[0156] Furthermore, the vehicle sudden stop monitoring module 301 specifically includes:

[0157] A video cutting unit, used for cutting the video images according to a preset interval time to obtain a video to be identified;

[0158] A key frame extraction unit, used to extract key frames from the video to be identified to obtain a key frame image;

[0159] The content recognition unit is used to input the key frame image into the content recognition model to determine whether there is a suddenly stopped vehicle in the video image.

[0160] Furthermore, the parking time acquisition module 302 specifically includes:

[0161] A frame number calculation unit, used to obtain the frame number information of the key frame image;

[0162] The parking time calculation unit is used to determine the parking time of the suddenly stopped vehicle based on the frame number information and the time axis of the video to be identified.

[0163] Furthermore, the vehicle trajectory identification module 303 specifically includes:

[0164] A video clip capture unit, used to capture a video clip within a predetermined time interval before the parking moment from the video image;

[0165] A video segment parsing unit, used for extracting video frames of the video segment;

[0166] The vehicle trajectory recognition unit is used to import the video frames into a pre-trained trajectory recognition model according to the time sequence of the video clips to obtain the vehicle trajectory of the suddenly stopped vehicle.

[0167] Furthermore, the vehicle trajectory identification unit specifically includes:

[0168] A feature extraction subunit, used to extract features from the video frame to obtain a feature sequence;

[0169] A vector conversion subunit, used for performing vector conversion on the feature sequence to obtain a feature vector sequence;

[0170] A vector concatenation subunit, used for concatenating the feature vector sequence to obtain an initial feature matrix;

[0171] A convolution operation subunit, used for performing a convolution operation on the initial feature matrix to obtain a convolution feature matrix;

[0172] The matrix combination subunit is used to combine the convolution feature matrices to obtain the vehicle trajectory of the suddenly stopped vehicle.

[0173] Furthermore, the accident responsibility determination module 306 specifically includes:

[0174] A related entity construction unit, used to construct a related entity of the accident vehicle based on the vehicle trajectory and the driving information;

[0175] The accident responsibility determination unit is used to import the associated entity into the traffic control regulations knowledge graph and output the accident responsibility of the accident vehicle.

[0176] Furthermore, the vehicle accident handling device also includes

[0177] A claim guidance module, used to generate a guidance interface for accident claim operations and display the guidance interface to the owner of the accident vehicle;

[0178] A claim settlement operation module, used to respond to the operation of the owner of the accident vehicle on the guidance interface and generate claim settlement operation steps, wherein the claim settlement operation steps at least include taking photos for evidence collection and damage assessment analysis;

[0179] A photo-taking and evidence-collecting module, used to generate a viewfinder, and according to the claim settlement operation steps on the guidance interface, instruct the owner of the accident vehicle to use the viewfinder to complete the photo-taking and evidence-collecting and upload the vehicle damage photos;

[0180] A damage assessment analysis module is used to identify the vehicle damage photo using a pre-trained vehicle damage recognition model to obtain the degree of vehicle damage;

[0181] The accident compensation module is used to generate a compensation plan based on the degree of vehicle damage and the accident liability.

[0182] In this embodiment, the present application discloses a vehicle accident handling device, which belongs to the field of artificial intelligence technology. The present application obtains video images of the monitored road section in real time, and determines whether there is a stopped vehicle in the video image. If there is a stopped vehicle in the image, the stopping time of the stopped vehicle is immediately determined, and a video clip within a predetermined time interval before the stopping time is intercepted from the video image, and the vehicle trajectory of the stopped vehicle is identified from the video clip, and it is determined whether there is an intersection point in the vehicle trajectory of the stopped vehicle. If there is an intersection point in the vehicle trajectory of the stopped vehicle, the stopped vehicle is determined to be an accident vehicle, and the driving information of the accident vehicle is obtained, and the accident responsibility of the accident vehicle is determined based on the vehicle trajectory and driving information using a pre-built traffic control regulations knowledge graph, the owner information of the accident vehicle is obtained, and the accident responsibility document is generated based on the owner information and the accident responsibility, and the evacuation guidance information is generated based on the vehicle trajectory, and the accident vehicle is instructed to evacuate the accident scene according to the evacuation guidance information. This application provides a process for handling minor vehicle accidents. By identifying the stopped vehicles and vehicle trajectories, and combining the knowledge graph of traffic regulations to automatically determine the accident responsibility of the accident vehicle, the efficiency of handling minor vehicle accidents is significantly improved, and traffic congestion caused by minor vehicle accidents is effectively solved. In addition, this application also provides an automatic claims settlement process for minor vehicle accidents, further improving the efficiency of handling minor vehicle accidents.

[0183] To solve the above technical problems, the present application also provides a computer device. Figure 4 , Figure 4 This is a basic structural block diagram of the computer device in this embodiment.

[0184] The computer device 4 includes a memory 41, a processor 42, and a network interface 43 that are interconnected through a system bus. It should be noted that the figure only shows a computer device 4 with components 41-43, but it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented instead. Among them, those skilled in the art can understand that the computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (Application Specific Integrated Circuit, ASIC), programmable gate arrays (Field-Programmable Gate Array, FPGA), digital processors (Digital Signal Processor, DSP), embedded devices, etc.

[0185] The computer device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The computer device may interact with a user through a keyboard, a mouse, a remote controller, a touch pad, or a voice control device.

[0186] The memory 41 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 41 can be an internal storage unit of the computer device 4, such as a hard disk or memory of the computer device 4. In other embodiments, the memory 41 can also be an external storage device of the computer device 4, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), etc. equipped on the computer device 4. Of course, the memory 41 can also include both the internal storage unit of the computer device 4 and its external storage device. In this embodiment, the memory 41 is generally used to store the operating system and various application software installed on the computer device 4, such as computer-readable instructions of the vehicle accident handling method, etc. In addition, the memory 41 can also be used to temporarily store various types of data that have been output or are to be output.

[0187] The processor 42 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chips in some embodiments. The processor 42 is generally used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is used to run computer-readable instructions or process data stored in the memory 41, such as computer-readable instructions for running the vehicle accident handling method.

[0188] The network interface 43 may include a wireless network interface or a wired network interface. The network interface 43 is generally used to establish a communication connection between the computer device 4 and other electronic devices.

[0189] The present application discloses a computer device, which belongs to the field of artificial intelligence technology. The present application obtains video images of a monitored road section in real time, and determines whether there is a stopped vehicle in the video image. If there is a stopped vehicle in the image, the stopping time of the stopped vehicle is immediately determined, and a video clip within a predetermined time interval before the stopping time is intercepted from the video image, and the vehicle trajectory of the stopped vehicle is identified from the video clip, and the vehicle trajectory of the stopped vehicle is determined to have an intersection point. If there is an intersection point in the vehicle trajectory of the stopped vehicle, the stopped vehicle is determined to be an accident vehicle, and the driving information of the accident vehicle is obtained, and the accident responsibility of the accident vehicle is determined based on the vehicle trajectory and driving information using a pre-built traffic control regulations knowledge graph, the owner information of the accident vehicle is obtained, and the accident responsibility document is generated based on the owner information and the accident responsibility, and the evacuation guidance information is generated based on the vehicle trajectory, and the accident vehicle is instructed to evacuate the accident scene according to the evacuation guidance information. The present application provides a processing flow for minor vehicle accidents, which significantly improves the efficiency of handling minor vehicle accidents by identifying the stopped vehicle and the vehicle trajectory, and automatically determines the accident responsibility of the accident vehicle in combination with the traffic control regulations knowledge graph, and effectively solves the traffic congestion caused by minor vehicle accidents. In addition, this application also provides an automatic claims process for minor vehicle accidents, further improving the efficiency of handling minor vehicle accidents.

[0190] The present application also provides another embodiment, namely, providing a computer-readable storage medium, wherein the computer-readable storage medium stores computer-readable instructions, and the computer-readable instructions can be executed by at least one processor to enable the at least one processor to perform the steps of the vehicle accident handling method as described above.

[0191] The present application discloses a storage medium, which belongs to the field of artificial intelligence technology. The present application obtains video images of the monitored road section in real time, and determines whether there is a stopped vehicle in the video image. If there is a stopped vehicle in the image, the stopping time of the stopped vehicle is immediately determined, and the video clips within a predetermined time interval before the stopping time are intercepted from the video image, and the vehicle trajectory of the stopped vehicle is identified from the video clips, and the vehicle trajectory of the stopped vehicle is determined to have an intersection point. If there is an intersection point in the vehicle trajectory of the stopped vehicle, the stopped vehicle is determined to be an accident vehicle, and the driving information of the accident vehicle is obtained, and the accident responsibility of the accident vehicle is determined based on the vehicle trajectory and driving information using a pre-built traffic control regulations knowledge graph, the owner information of the accident vehicle is obtained, and the accident responsibility document is generated based on the owner information and the accident responsibility, and the evacuation guidance information is generated based on the vehicle trajectory, and the accident vehicle is instructed to evacuate the accident scene according to the evacuation guidance information. The present application provides a processing flow for minor vehicle accidents, which significantly improves the efficiency of handling minor vehicle accidents by identifying the stopped vehicle and the vehicle trajectory, and automatically determines the accident responsibility of the accident vehicle in combination with the traffic control regulations knowledge graph, and effectively solves the traffic congestion caused by minor vehicle accidents. In addition, this application also provides an automatic claims process for minor vehicle accidents, further improving the efficiency of handling minor vehicle accidents.

[0192] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0193] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0194] Obviously, the embodiments described above are only some embodiments of the present application, rather than all embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application is described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific implementation methods, or to perform equivalent replacement of some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of this application, directly or indirectly used in other related technical fields, is similarly within the scope of patent protection of this application.

Claims

1. A vehicle accident handling method, characterized in that: include: Acquire video images of the monitored road section in real time, and determine whether there is a suddenly stopped vehicle in the video images based on a pre-trained content recognition model; If there is a suddenly stopped vehicle in the video image, determining the stopping time of the suddenly stopped vehicle based on the time axis of the video image; intercepting a video clip within a predetermined time interval before the stopping moment from the video image, and identifying a vehicle trajectory of the suddenly stopped vehicle from the video clip; Determining whether there is an intersection point in the vehicle trajectory of the suddenly stopped vehicle; If the vehicle trajectory of the suddenly stopped vehicle has an intersection, then the suddenly stopped vehicle is determined to be an accident vehicle; Acquire driving information of the accident vehicle, and determine the accident responsibility of the accident vehicle using a pre-built traffic control regulations knowledge graph based on the vehicle trajectory and the driving information; Acquiring the owner information of the accident vehicle, and generating an accident responsibility document based on the owner information and the accident responsibility; Evacuation guidance information is generated based on the vehicle trajectory, and the accident vehicle is instructed to evacuate the accident scene according to the evacuation guidance information.

2. The vehicle accident handling method according to claim 1, characterized in that: The step of acquiring the video image of the monitored road section in real time and judging whether there is a suddenly stopped vehicle in the video image based on a pre-trained content recognition model specifically includes: Cutting the video images according to a preset time interval to obtain a video to be identified; Extracting key frames from the video to be identified to obtain key frame images; The key frame image is input into the content recognition model to determine whether there is a vehicle that suddenly stops in the video image.

3. The vehicle accident handling method according to claim 2, characterized in that: The step of determining the stopping time of the suddenly stopped vehicle based on the time axis of the video image specifically includes: Obtaining frame number information of the key frame image; The stopping time of the suddenly stopped vehicle is determined based on the frame number information and the time axis of the video to be identified.

4. The vehicle accident handling method according to claim 1, characterized in that: The step of capturing a video clip within a predetermined time interval before the stopping moment from the video image, and identifying the vehicle trajectory of the suddenly stopped vehicle from the video clip specifically includes: Extracting a video clip within a predetermined time interval before the parking moment from the video image; Extracting video frames of the video clip; The video frames are imported into a pre-trained trajectory recognition model according to the time sequence of the video clips to obtain the vehicle trajectory of the suddenly stopped vehicle.

5. The vehicle accident handling method according to claim 4, characterized in that: The step of importing the video frames into a pre-trained trajectory recognition model according to the time sequence of the video clips to obtain the vehicle trajectory of the suddenly stopped vehicle specifically includes: Extracting features from the video frame to obtain a feature sequence; Performing vector transformation on the feature sequence to obtain a feature vector sequence; splicing the feature vector sequence to obtain an initial feature matrix; Performing a convolution operation on the initial feature matrix to obtain a convolution feature matrix; The convolution feature matrices are combined to obtain the vehicle trajectory of the suddenly stopped vehicle.

6. The vehicle accident handling method according to any one of claims 1 to 5, characterized in that: The step of determining the accident responsibility of the accident vehicle using a pre-built traffic control regulations knowledge graph based on the vehicle trajectory and the driving information specifically includes: Constructing an associated entity of the accident vehicle based on the vehicle trajectory and the driving information; The associated entity is imported into the traffic control regulations knowledge graph, and the accident responsibility of the accident vehicle is output.

7. The vehicle accident handling method according to claim 6, characterized in that: After the step of obtaining the owner information of the accident vehicle and generating an accident responsibility document based on the owner information and the accident responsibility, the method further includes: Generating a guidance interface for accident claim settlement operations, and displaying the guidance interface to the owner of the accident vehicle; In response to the operation of the owner of the accident vehicle on the guidance interface, generating claim settlement operation steps, wherein the claim settlement operation steps at least include taking photos for evidence collection and damage assessment analysis; Generate a framing frame, and instruct the owner of the accident vehicle to use the framing frame to complete photo collection and upload vehicle damage photos according to the claim settlement operation steps on the guidance interface; Using a pre-trained vehicle damage recognition model to recognize the vehicle damage photo, and obtaining the degree of vehicle damage; A compensation plan is generated based on the degree of vehicle damage and the accident liability.

8. A vehicle accident handling device, characterized in that: include: The vehicle sudden stop monitoring module is used to obtain the video image of the monitored road section in real time and determine whether there is a sudden stop vehicle in the video image based on a pre-trained content recognition model; A parking time acquisition module, used for determining the parking time of the suddenly stopped vehicle based on the time axis of the video image when there is a suddenly stopped vehicle in the video image; A vehicle track recognition module, used to capture a video clip within a predetermined time interval before the stopping moment from the video image, and recognize the vehicle track of the suddenly stopped vehicle from the video clip; A track intersection judgment module is used to judge whether there is an intersection point in the vehicle track of the suddenly stopped vehicle; An accident vehicle confirmation module, used for determining that the suddenly stopped vehicle is an accident vehicle if there is an intersection in the vehicle trajectory of the suddenly stopped vehicle; An accident responsibility determination module, used to obtain the driving information of the accident vehicle, and determine the accident responsibility of the accident vehicle based on the vehicle trajectory and the driving information using a pre-built traffic control regulations knowledge graph; A responsibility document generation module, used to obtain the owner information of the accident vehicle, and generate an accident responsibility document based on the owner information and the accident responsibility; The vehicle evacuation guidance module is used to generate evacuation guidance information based on the vehicle trajectory and instruct the accident vehicle to evacuate the accident scene according to the evacuation guidance information.

9. A computer device, characterized in that: The method comprises a memory and a processor, wherein the memory stores computer-readable instructions, and the processor implements the steps of the vehicle accident handling method as claimed in any one of claims 1 to 7 when executing the computer-readable instructions.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by a processor, the steps of the vehicle accident handling method according to any one of claims 1 to 7 are implemented.

Images