Multi-model interconnected highway event detection method and device and storage medium

Abstract

The invention discloses a multi-model interconnected highway event detection method and device and a storage medium, and can be applied to the technical field of highway event detection. According to the invention, after the secondary rechecking unit, the expressway detection server, the expressway event monitoring management system and the expressway video monitoring service system are respectively arranged, the expressway detection server analyzes the video data collected by at least one expressway camera; the second-level rechecking unit is used for rechecking the first expressway event obtained through analysis, and after a second expressway event obtained through rechecking is sent to the expressway event monitoring and management system, overall management is carried out on the second expressway event through the expressway event monitoring and management system; and the overall management result is sent to an expressway video monitoring service system, so that expressway event prompting is carried out through the expressway video monitoring service system, and therefore, the detection accuracy of expressway events can be effectively improved.

Description

Technical Field

[0001] The present application relates to the technical field of highway event detection, and in particular to a highway event detection method, device and storage medium with multi-model interconnection. Background Art

[0002] In related technologies, with the development of the highway network, the passenger flow on highways continues to increase, and the traffic safety accidents caused by them are also increasing. The current method for detecting highway incidents is to detect them through MEC devices deployed at the edge. However, due to the performance of MEC devices, the computing power of the neural network algorithm deployed in the MEC devices is also poor, resulting in low accuracy in detecting highway incidents.

[0003] In summary, the technical problems existing in the relevant technologies need to be improved. Summary of the invention

[0004] The main purpose of the embodiments of the present application is to propose a multi-model interconnected highway event detection method, device and storage medium, which can effectively improve the detection accuracy of highway events.

[0005] To achieve the above-mentioned purpose, an embodiment of the present application proposes a multi-model interconnected highway event detection method, the method is applied to a secondary review unit, the secondary review unit interacts with a highway detection server and a highway event monitoring management system respectively, the highway event monitoring management system interacts with a highway video surveillance business system, and the method includes the following steps:

[0006] Acquire a first highway event sent by at least one highway detection server; each of the highway detection servers is used to analyze video data collected by at least one highway camera to obtain the first highway event;

[0007] Reviewing the first highway event to obtain a second highway event;

[0008] The second highway event is sent to the highway event monitoring and management system; the highway event monitoring and management system is used to coordinate the management of the second highway event to obtain a coordinated management result, and send the coordinated management result to the highway video surveillance business system.

[0009] In some embodiments, analyzing the video data collected by at least one highway camera to obtain the first highway event includes:

[0010] Obtaining video data collected by at least one highway camera;

[0011] Performing highway event recognition on the video data by using a first neural network model to obtain a third highway event;

[0012] Calculating the confidence level of the third highway event;

[0013] The first highway event is obtained by screening the third highway events according to the relationship between the confidence level and a preset threshold.

[0014] In some embodiments, reviewing the first highway event to obtain the second highway event includes:

[0015] The first highway event is identified by a second neural network model to obtain the second highway event; wherein the model parameters of the second neural network model are greater than the model parameters of the first neural network model.

[0016] In some embodiments, the step of performing overall management on the second highway event to obtain an overall management result includes:

[0017] Obtaining vehicle speed information of the second highway event;

[0018] Collecting slow vehicle information according to the vehicle speed information;

[0019] Obtaining the event type of the second highway event;

[0020] generating combined alarm information of the second highway event according to the event type;

[0021] The overall management result is generated according to the speed vehicle information, the combined alarm information, the alarm push information or the confidence level corresponding to the second highway event.

[0022] In some embodiments, the highway detection server further interacts with an AI capability middle station, which is deployed in the cloud and is used to perform the following steps:

[0023] Updating a first preset training set according to the first highway event;

[0024] Performing update training on the third neural network model according to the updated first preset training set to obtain a fourth neural network model;

[0025] The fourth neural network model is sent to the highway detection server; and the highway detection server updates the first neural network model according to the fourth neural network model.

[0026] In some embodiments, the secondary review unit includes a large model integrated machine or a highway event review server, the large model integrated machine is deployed in the cloud, and the highway event review server is deployed on the edge side corresponding to the highway detection server.

[0027] In some embodiments, the AI ​​capability platform is further used to perform the following steps:

[0028] updating a second preset training set according to the second highway event;

[0029] Performing update training on the fifth neural network model according to the updated second preset training set to obtain a sixth neural network model;

[0030] The sixth neural network model is sent to the secondary review unit; and the secondary review unit updates the second neural network model according to the sixth neural network model.

[0031] To achieve the above-mentioned purpose, another aspect of the embodiment of the present application proposes a multi-model interconnected highway event detection device, the device is applied to a secondary review unit, the secondary review unit interacts with a highway detection server and a highway event monitoring management system respectively, the highway event monitoring management system interacts with a highway video surveillance business system, and the device includes:

[0032] The first module is used to obtain a first highway event sent by at least one highway detection server; each of the highway detection servers is used to analyze video data collected by at least one highway camera to obtain the first highway event;

[0033] The second module is used to review the first highway event to obtain a second highway event;

[0034] The third module is used to send the second highway event to the highway event monitoring and management system; the highway event monitoring and management system is used to coordinate the management of the second highway event to obtain a coordinated management result, and send the coordinated management result to the highway video surveillance business system.

[0035] To achieve the above objective, another aspect of the present application provides a computer device, including:

[0036] at least one processor;

[0037] at least one memory for storing at least one program;

[0038] When the at least one program is executed by the at least one processor, the at least one processor implements the above method.

[0039] To achieve the above objective, another aspect of an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program implements the above method when executed by a processor.

[0040] The embodiments of the present application include at least the following beneficial effects: The present application provides a multi-model interconnected highway event detection method, device and storage medium. The scheme respectively sets up a secondary review unit, a highway detection server, a highway event monitoring and management system and a highway video surveillance business system, analyzes the video data collected by at least one highway camera through the highway detection server to obtain a first highway event, reviews the first highway event through the secondary review unit to obtain a second highway event, and then sends the second highway event to the highway event monitoring and management system. The second highway event is comprehensively managed by the highway event monitoring and management system to obtain a comprehensive management result, and the comprehensive management result is sent to the highway video surveillance business system, so that the highway event prompt is performed through the highway video surveillance business system, thereby realizing the multi-level recognition and detection function of highway events, and effectively improving the detection accuracy of highway events. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a schematic diagram of an interactive scenario of highway incident detection in the prior art;

[0042] Figure 2 is a flow chart of a highway event detection method with multi-model interconnection provided in an embodiment of the present application;

[0043] Figure 3 It is a schematic diagram of an interactive scenario of highway event detection provided by an embodiment of the present application;

[0044] Figure 4 It is a structural schematic diagram of a highway event detection device with multiple models interconnected provided in an embodiment of the present application;

[0045] Figure 5 It is a schematic diagram of the hardware structure of the computer device provided in the embodiment of the present application. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail below in conjunction with the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the embodiments of the present application, they are only examples of devices and methods consistent with some aspects of the embodiments of the present application.

[0047] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0048] Before describing the embodiments of the present application in detail, some nouns and terms involved in the embodiments of the present application are first described. The nouns and terms involved in the embodiments of the present application are subject to the following interpretations:

[0049] Artificial Intelligence (AI) is a new technical science that studies and develops theories, methods, technologies and application systems for simulating, extending and expanding human intelligence. AI is a branch of computer science that attempts to understand the essence of intelligence and produce a new type of intelligent machine that can respond in a similar way to human intelligence. Research in this field includes robots, language recognition, image recognition, natural language processing and expert systems. AI can simulate the information process of human consciousness and thinking. AI is also a 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.

[0050] Mobile Edge Computing (MEC) is the product of I CT integration, and has become a key technology to support operators in their 5G network transformation to meet the development needs of services such as high-definition video, VR / AR, industrial Internet, and Internet of Vehicles.

[0051] In the related art, highway events refer to events that occur non-periodically on closed highways and may lead to a decrease in road capacity, traffic safety accidents, or abnormal increase in traffic demand, such as vehicle breakdown, illegal parking, spilled objects, driving in the wrong direction, intrusion of non-motor vehicles or pedestrians, etc. Since vehicles on highways are in a high-speed driving state, the type, time and location of highway events are unpredictable. Therefore, once a traffic event occurs, it is very easy to cause serious traffic congestion and driving delays. If it is not eliminated in time, it is easy to cause subsequent secondary accidents, which seriously hinders the normal operation of the highway system. Therefore, highway event detection is an important part of highway operation and maintenance.

[0052] like Figure 1 As shown in the figure, the existing technology mainly deploys the trained video event detection algorithm (such as the target detection algorithm based on deep neural network) to the highway event detection server, and deploys the server to the section data center close to the highway monitoring camera. Therefore, the highway event detection server can be understood as an MEC device deployed at the edge. Through the video streaming method, a MEC device accesses multiple groups of video streams. The built-in event detection algorithm of MEC automatically identifies some predefined abnormal events in the video stream, and pushes the event detection results to the business system of the monitoring center (such as the highway event monitoring and management system), and the monitor reviews the events identified by MEC.

[0053] In reality, in order to pursue the advantage of road unit price (monitoring unit price of each video channel), low-performance MEC devices are usually set up, resulting in relatively weak neural network algorithms. As a result, the event detection capture rate (recall rate) and accuracy are relatively low, resulting in a large number of important events being missed and a large number of false alarms, which actually cannot solve the problem of intelligent detection of highway events.

[0054] In view of this, a multi-model interconnected highway event detection method, device and storage medium are provided in the embodiments of the present application. After respectively setting up a secondary review unit, a highway detection server, a highway event monitoring and management system and a highway video surveillance business system, the embodiment of the present application analyzes the video data collected by at least one highway camera through the highway detection server to obtain a first highway event, reviews the first highway event through the secondary review unit to obtain a second highway event, and then sends the second highway event to the highway event monitoring and management system. The second highway event is comprehensively managed by the highway event monitoring and management system to obtain a comprehensive management result, and the comprehensive management result is sent to the highway video surveillance business system, so that the highway event prompt is performed through the highway video surveillance business system, thereby realizing the multi-level recognition and detection function of highway events, and effectively improving the detection accuracy of highway events.

[0055] The multi-model interconnected highway event detection method provided in the embodiment of the present application relates to the field of highway event detection technology. The multi-model interconnected highway event detection method provided in the embodiment of the present application can be applied to a terminal, can also be applied to a server, and can also be software running in a terminal or a server. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, and a car terminal, etc., but is not limited to this; the server side can be configured as an independent physical server, or it can be configured as a server cluster or distributed system composed of multiple physical servers, and can also be configured as 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, CDN, and big data and artificial intelligence platforms. The server can also be a node server in a blockchain network; the software can be an application of a multi-model interconnected highway event detection method, etc., but is not limited to the above forms.

[0056] 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.

[0057] In the embodiments of the present application, Figure 2 The method shown can also be applied to Figure 3 In the interactive scenario shown in Figure 3 As shown, the interactive scenario includes several highway cameras, a highway detection server (MEC device), a secondary review unit, a highway event monitoring management system, and a highway video surveillance business system. It can be understood that each highway detection server can be connected to at least one highway camera, each secondary review unit can be connected to the highway event monitoring management system and at least one highway detection server, respectively, and the highway event monitoring management system is connected to the highway video surveillance business system.

[0058] by Figure 2 The method shown is applied to Figure 3 Take the secondary review unit of as an example, Figure 2 The method shown includes but is not limited to steps S210 to S230:

[0059] Step S210: obtaining a first highway event sent by at least one highway detection server; wherein each highway detection server is used to analyze video data collected by at least one highway camera to obtain the first highway event;

[0060] Step S220, reviewing the first expressway event to obtain a second expressway event;

[0061] Step S230: Send the second expressway event to the expressway event monitoring and management system, so that the expressway event monitoring and management system performs overall management on the second expressway event to obtain an overall management result, and sends the overall management result to the expressway video monitoring business system.

[0062] In the embodiment of the present application, the highway detection server is deployed on the road section side, and is used to access the video data of the highway monitoring camera and perform preliminary screening of the highway events. It can be understood that in the preliminary screening process, after the highway detection server obtains the video data collected by at least one highway camera, the highway event recognition is performed on the video data through the first neural network model to obtain the third highway event, and the confidence of each third highway event is calculated; when the confidence exceeds the upper threshold of the threshold range, it is determined that the corresponding third highway event has a high degree of credibility, and the third highway event is used as the detection result of the highway event; when the confidence exceeds the lower threshold of the threshold range (preset threshold), the corresponding third highway event may be misjudged. In order to ensure that no missed detection is detected, this embodiment will take the corresponding third highway event as the first highway event and transfer it to the secondary review unit for review; when the confidence is less than the lower threshold of the threshold range, it can be considered that the detection is normal and no event occurs. In the embodiment of the present application, the first neural network model deployed by the highway detection server can be set as a network model with a high capture rate (recall rate), so as to ensure that abnormal highway events will not be missed.

[0063] In an embodiment of the present application, the first highway event obtained by preliminary screening through the highway detection server may have a large number of false positive linearity, and this embodiment is reviewed by a secondary review unit. It can be understood that in this embodiment, the first highway event obtained by preliminary screening by the highway detection server is input into the second neural network model, so as to obtain the second highway event by performing highway event recognition on the first highway event through the second neural network model. It can be understood that the model parameters of the second neural network model of this embodiment are greater than the model parameters of the first neural network model, so that the secondary review unit is a computing power server with excellent computing power, which can achieve a high-accuracy abnormal event recognition effect, so that the first highway event that may be misjudged can be checked twice to improve the accuracy of event detection.

[0064] Specifically, the first neural network model can be a special recognition model built by the Yolo series, and its model parameters are between millions of parameters and tens of millions of parameters. Among them, the highway detection server deployed on the road section can simultaneously screen a batch of high-speed highway video data in real time. The first neural network model is characterized by high discrimination efficiency but low confidence in the discrimination of complex events. The parameters of the second neural network model are between tens of billions of parameters and hundreds of billions of parameters. The characteristics of this neural network model are large parameters and high requirements for the operating environment, but the accuracy of the discrimination results is high, which is suitable for secondary verification of events with low confidence in the first neural network model.

[0065] In the embodiment of the present application, the second highway event reviewed by the secondary review unit is pushed to the highway event monitoring and management system for overall management. Among them, the highway event monitoring and management system provides event configuration functions: event definition, the user can define the "slow driving" event as the vehicle speed is lower than XX% of the average speed of the traffic flow in the section; merge forwarding, the user can merge and forward the abnormal events of the system alarm according to the rules, for example, "the same event is continuously alarmed x times and then merged into one alarm."; Maximum number of pushes: refers to the system pushing the same alarm up to X times within the specified time t; event identification threshold is the condition for the system to generate an alarm push. If the event confidence exceeds the threshold, it is considered that the event detection is highly credible, and the event alarm is pushed to the highway video surveillance business system, otherwise it will not be pushed. After configuring the event of the highway event monitoring and management system, this embodiment can coordinate the management of the second highway event to obtain a coordinated management result. It can be understood that the coordinated management process of this embodiment can be achieved by obtaining the vehicle speed information of the second highway event, and counting the slow vehicle information based on the vehicle speed information; and obtaining the event type of the second highway event, and generating combined alarm information for the second highway event based on the event type; and then generating a coordinated management result based on the speed vehicle information, combined alarm information, alarm push information or the confidence level corresponding to the second highway event.

[0066] In the embodiment of the present application, the highway detection server also interacts with the AI ​​capability middle station. The AI ​​capability middle station is deployed in the cloud capability middle station of the highway group, and is used to update the first preset training set according to the first highway event, update and train the third neural network model according to the updated first preset training set to obtain a fourth neural network model, and send the fourth neural network model to the highway detection server, so that the highway detection server updates the first neural network model according to the fourth neural network model, thereby realizing the online upgrade of the first neural network model in the highway detection server.

[0067] It can be understood that the secondary review unit also interacts with the AI ​​capability middle station, and the AI ​​capability middle station can perform online upgrade of the second neural network model in the secondary review unit. The upgrade process is to update the second preset training set according to the second highway event, and then update the fifth neural network model according to the updated second preset training set to obtain the sixth neural network model, and then send the sixth neural network model to the secondary review unit so that the secondary review unit updates the second neural network model according to the sixth neural network model.

[0068] In an embodiment of the present application, the secondary review unit includes a large model integrated machine or a highway event review server, the large model integrated machine is deployed in the cloud, and the highway event review server is deployed on the edge side corresponding to the highway detection server. During the application process, the large model integrated machine or the highway event review server can be selected according to the actual selection of the user to review the first highway event. Among them, the highway event review server has the characteristics of fast review speed, high efficiency, low latency but high cost during the application process, and the large model integrated machine has the characteristics of low cost but long latency. When the large model integrated machine is selected for event review, it can interact with the highway detection server and the highway event monitoring and management system through the API interface corresponding to the large model integrated machine.

[0069] It can be seen from this that the method of the embodiment of the present application can economically achieve efficient and accurate detection of highway events through a combination of three-level capabilities of a highway detection server, a dimethyl review unit and an AI capability center; and achieve decoupled management of events through a highway event monitoring and management module, thereby achieving decoupling from the highway detection server, and being able to freely connect to other unrestricted brand highway detection servers, and also achieving decoupling from the business system to avoid system event push affecting business system operation.

[0070] Reference Figure 4 The embodiment of the present application provides a multi-model interconnected highway event detection device, which is applied to a secondary review unit, which interacts with a highway detection server and a highway event monitoring management system respectively, and the highway event monitoring management system interacts with a highway video surveillance business system. Figure 4 As shown, the device comprises:

[0071] The first module is used to obtain a first highway event sent by at least one highway detection server; wherein each highway detection server is used to analyze video data collected by at least one highway camera to obtain the first highway event;

[0072] The second module is used to review the first highway event to obtain a second highway event;

[0073] The third module is used to send the second expressway event to the expressway event monitoring and management system; wherein the expressway event monitoring and management system is used to coordinate the management of the second expressway event to obtain a coordinated management result, and send the coordinated management result to the expressway video surveillance business system.

[0074] It can be understood that the contents of the above method embodiments are all applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0075] The embodiment of the present application also provides a computer device, the computer device includes a memory and a processor, the memory stores a computer program, and the processor implements the above-mentioned multi-model interconnected highway event detection method when executing the computer program. The computer device can be any intelligent terminal including a tablet computer, a car computer, etc.

[0076] It can be understood that the contents of the above method embodiments are all applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0077] See also Figure 5 , Figure 5 The hardware structure of a computer device according to another embodiment is shown, and the computer device includes:

[0078] The processor 510 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application.

[0079] The memory 520 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 520 can store an operating system and other application programs. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 520, and the processor 510 calls and executes the highway event detection method with multi-model interconnection in the embodiment of this application;

[0080] Input / output interface 530, used to implement information input and output;

[0081] Communication interface 540, used to realize communication interaction between the device and other devices, which can be realized through wired mode (such as USB, network cable, etc.) or wireless mode (such as mobile network, WI FI, Bluetooth, etc.);

[0082] bus 550 , which transmits information between the various components of the device (e.g., processor 510 , memory 520 , input / output interface 530 , and communication interface 540 );

[0083] The processor 510 , the memory 520 , the input / output interface 530 and the communication interface 540 are connected to each other in communication within the device via the bus 550 .

[0084] An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the above-mentioned multi-model interconnected highway event detection method.

[0085] It can be understood that the contents of the above method embodiments are all applicable to the present storage medium embodiments, the functions specifically implemented by the present storage medium embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0086] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0087] The embodiments described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0088] Those skilled in the art will appreciate that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0089] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0090] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0091] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0092] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0093] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the above units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0094] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0095] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0096] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including multiple instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store programs.

[0097] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but the scope of the rights of the present invention is not limited thereto. Any modification, equivalent substitution and improvement made by a person skilled in the art without departing from the scope and essence of the present invention should be within the scope of the rights of the present invention.

Claims

1. A highway incident detection method with multiple interconnected models, characterized in that: The method is applied to a secondary review unit, the secondary review unit interacts with a highway detection server and a highway event monitoring management system respectively, the highway event monitoring management system interacts with a highway video monitoring business system, and the method includes the following steps: Acquire a first highway event sent by at least one highway detection server; each of the highway detection servers is used to analyze video data collected by at least one highway camera to obtain the first highway event; Reviewing the first highway event to obtain a second highway event; The second highway event is sent to the highway event monitoring and management system; the highway event monitoring and management system is used to coordinate the management of the second highway event to obtain a coordinated management result, and send the coordinated management result to the highway video surveillance business system.

2. The method according to claim 1, characterized in that The step of analyzing the video data collected by at least one highway camera to obtain the first highway event includes: Obtaining video data collected by at least one highway camera; Performing highway event recognition on the video data by using a first neural network model to obtain a third highway event; Calculating the confidence level of the third highway event; The first highway event is obtained by screening the third highway events according to the relationship between the confidence level and a preset threshold.

3. The method according to claim 2, characterized in that The reviewing of the first highway event to obtain a second highway event includes: The first highway event is identified by a second neural network model to obtain the second highway event; wherein the model parameters of the second neural network model are greater than the model parameters of the first neural network model.

4. The method according to claim 1, characterized in that The coordinated management of the second highway incident to obtain a coordinated management result includes: Obtaining vehicle speed information of the second highway event; Collecting slow vehicle information according to the vehicle speed information; Obtaining the event type of the second highway event; generating combined alarm information of the second highway event according to the event type; The overall management result is generated according to the speed vehicle information, the combined alarm information, the alarm push information or the confidence level corresponding to the second highway event.

5. The method according to claim 3, characterized in that: The highway detection server also interacts with an AI capability middle station, which is deployed in the cloud and is used to perform the following steps: Updating a first preset training set according to the first highway event; Performing update training on the third neural network model according to the updated first preset training set to obtain a fourth neural network model; The fourth neural network model is sent to the highway detection server; and the highway detection server updates the first neural network model according to the fourth neural network model.

6. The method according to claim 5, characterized in that The secondary review unit includes a large model integrated machine or a highway event review server. The large model integrated machine is deployed in the cloud, and the highway event review server is deployed on the edge side corresponding to the highway detection server.

7. The method according to claim 5, characterized in that The AI ​​capability platform is also used to perform the following steps: updating a second preset training set according to the second highway event; Performing update training on the fifth neural network model according to the updated second preset training set to obtain a sixth neural network model; The sixth neural network model is sent to the secondary review unit; and the secondary review unit updates the second neural network model according to the sixth neural network model.

8. A highway event detection device with multiple interconnected models, characterized in that: The device is applied to a secondary review unit, the secondary review unit interacts with a highway detection server and a highway event monitoring management system respectively, the highway event monitoring management system interacts with a highway video monitoring business system, and the device includes: The first module is used to obtain a first highway event sent by at least one highway detection server; each of the highway detection servers is used to analyze video data collected by at least one highway camera to obtain the first highway event; The second module is used to review the first highway event to obtain a second highway event; The third module is used to send the second highway event to the highway event monitoring and management system; the highway event monitoring and management system is used to coordinate the management of the second highway event to obtain a coordinated management result, and send the coordinated management result to the highway video surveillance business system.

9. A computer device, characterized in that: include: at least one processor; at least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing 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 7 is implemented.

Images