Intelligent early warning system and method for trunk road traffic safety based on edge computing

By deploying an intelligent traffic safety warning system based on edge computing on trunk roads, problems such as strong network dependence and large delay in video centralized surveillance technology have been solved, real-time traffic monitoring and early warning have been realized, and emergency response efficiency and traffic capacity have been improved. Reduced costs.

CN118658307BActive Publication Date: 2024-11-05ZHONGLU KEYUN (BEIJING) TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411141858.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-20
Publication Date
2024-11-05
Estimated Expiration
2044-08-20

AI Technical Summary

Technical Problem

The existing centralized video surveillance technology has problems such as strong network dependence, large delay, poor system expansion, high single-point failure risk and high cost in ordinary national and provincial trunk roads, making it difficult to achieve real-time traffic safety control and efficient emergency response.

Method used

Adopt the intelligent early warning system for trunk road traffic safety based on edge computing, including power supply facilities, network facilities, traffic perception facilities, traffic control and induction facilities and edge intelligent early warning devices, and automatically collect, process and make decisions on road traffic information through edge computing module. , generate and issue real-time warning policies to reduce data transmission delays and improve system scalability.

Benefits of technology

Realize instant monitoring of traffic conditions and rapid identification of abnormalities, shorten the time from detection to response, improve the efficiency and accuracy of traffic safety emergency response, reduce system construction and operation costs, and improve road operation efficiency and traffic capacity. .

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118658307B_ABST
    Figure CN118658307B_ABST
Patent Text Reader

Abstract

The present invention discloses an intelligent early warning system and method for traffic safety of trunk roads based on edge computing, which relates to the field of road traffic technology. The intelligent early warning system for traffic safety of trunk roads based on edge computing includes: a power supply facility unit, a network facility unit, a traffic perception facility unit, a traffic control and induction facility unit, and an edge intelligent early warning device unit. By deploying edge computing technology along the trunk roads, the present invention can directly process and analyze data near the data source, reduce data transmission delays, realize real-time monitoring of traffic conditions and rapid identification of abnormalities, effectively shorten the time from detection to response, and greatly improve the efficiency and accuracy of traffic safety emergency response.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of road traffic technology, and in particular to an intelligent early warning system and method for trunk road traffic safety based on edge computing. Background Art

[0002] As an indispensable transportation link between provinces and cities, ordinary national and provincial trunk roads not only bear heavy transportation tasks, but also play a core role in promoting economic development and facilitating transportation convenience. However, they face multiple challenges in operation: changeable road traffic and climate conditions, complex driving and passenger behaviors, and poor information flow, which lead to frequent traffic congestion and accidents, and thus cause certain losses. In order to meet these challenges, improving the traffic safety and traffic efficiency of ordinary national and provincial trunk roads has become an important issue that needs to be solved urgently.

[0003] The development of information technology and intelligent transportation technology has brought new space for improving the safety level and traffic capacity of ordinary national and provincial trunk roads. However, due to the limitations of road traffic conditions and construction funds, it is difficult for ordinary national and provincial trunk roads to carry out large-scale information construction and intelligent transportation construction like expressways. Therefore, for a long time, the means of traffic safety protection for ordinary national and provincial trunk roads have been very limited, and the development of its safety protection measures has mainly gone through the following three stages:

[0004] 1) Passive safety stage: This stage mainly relies on passive measures such as adding warning signs and installing guardrails to improve traffic safety. There is a lack of effective information management methods, and highway management personnel can only understand traffic conditions through patrols, alarms, etc. Although these passive traffic safety measures can reduce the risk of accidents and reduce accident losses to a certain extent, they cannot actively prevent traffic accidents, especially secondary accidents.

[0005] 2) Remote monitoring stage: In this stage, remote monitoring is gradually introduced to strengthen traffic safety. Surveillance cameras, variable information boards and other equipment are deployed on key roads, and videos are transmitted to provincial and municipal road network monitoring centers for centralized management. Center managers poll and view camera videos. Once abnormal traffic conditions are found, warning information is issued through variable information boards to provide route guidance and early warning to drivers and passengers. Traffic safety at this stage mainly depends on the density of surveillance cameras, and is also affected by the experience and judgment of managers, resulting in limited effect in improving traffic safety and traffic efficiency.

[0006] 3) Intelligent control stage: With the promotion and construction of smart highways, advanced technologies such as artificial intelligence and big data analysis have been more widely used in the field of traffic control. At the same time, the continuous rise and development of edge computing technology has provided the traffic system with more powerful data processing and decision-making capabilities, greatly expanding the application scenarios and scope of traffic control. Based on edge computing devices and intelligent algorithms, traffic data can be collected and analyzed in real time and accurately, real-time decision-making and response can be provided, traffic safety warnings can be issued autonomously, traffic congestion and accident rates can be reduced, and road safety can be improved. At this stage, the traffic system has the ability to learn and adapt autonomously, achieving more efficient and safer traffic system management.

[0007] The development of traffic safety measures aims to improve traffic efficiency, reduce congestion and increase traffic safety. With the continuous innovation of technology, traffic control is gradually developing in the direction of intelligence and automation. The traffic system can better cope with the growing travel demand and provide vehicles with safer, more convenient and efficient traffic services.

[0008] At present, video centralized monitoring technology has been widely used in ordinary national and provincial trunk roads and has achieved certain results. This technology realizes the monitoring and efficient control of traffic safety by centrally managing and monitoring multiple cameras and issuing early warning information in combination with variable information boards. Monitoring cameras are deployed along the highway and transmit video signals to provincial and municipal road network centers through dedicated networks or operator networks for centralized storage, playback and remote control, which greatly improves monitoring efficiency and management level and provides strong technical support for improving highway traffic safety. Fig.16 As shown in the figure, the overall architecture of this technology mainly includes two core parts: video acquisition and central management, which together constitute this efficient and intelligent traffic safety management system.

[0009] 1) Video acquisition: It consists of video surveillance equipment and a dedicated video network. The video surveillance equipment deployed on the side of the highway captures the road traffic situation in real time, and transmits the image and video data to the central management system through the dedicated video network. Video acquisition can cover roads in different areas, capture real-time traffic conditions and events, and provide data support for the center to conduct real-time monitoring, video storage and remote control.

[0010] 2) Central management: It is composed of the provincial and municipal road network center video management platform, which centrally manages the accessed videos and can receive, analyze and store video data from various monitoring devices. Central management generally includes two modes: manual and intelligent. In the manual mode, the administrator polls the video images of each monitoring point through the monitoring screen, manually judges abnormal traffic conditions, and performs remote control and command. In the intelligent mode, the central management platform has powerful data processing and analysis capabilities, and can intelligently analyze and identify traffic violations, abnormal events, etc., and conduct early warning processing after manual confirmation.

[0011] Although video centralized monitoring technology has achieved certain results in improving the level of highway traffic safety, it still has some limitations:

[0012] 1) Strong network dependence: Video centralized monitoring technology relies on the network to transmit large amounts of video data to the central management system for processing and analysis, so it has high requirements for network stability and bandwidth.

[0013] 2) Large delay: Since video data needs to be transmitted to the central management system before it can be processed and analyzed, and it is highly dependent on manual management and has low efficiency, it will lead to large event processing delays, which is not conducive to real-time traffic safety control and emergency response.

[0014] 3) Poor system scalability: The scalability of centralized video surveillance technology is limited by the processing power and storage capacity of the central management system. When the monitoring scope expands, the system equipment needs to be continuously upgraded to meet the needs.

[0015] 4) Single point failure has a wide impact range: Video centralized monitoring technology has the risk of a single failure point affecting the entire system. Once the backbone network or central management system fails, the entire monitoring system will be paralyzed.

[0016] 5) High cost: Deploying and maintaining a centralized video surveillance system requires a large amount of capital investment, including costs for network construction, video management platform construction, system operation and maintenance, etc. This creates a great economic pressure on some resource-scarce regions.

[0017] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention

[0018] In response to the problems in the related technology, the present invention proposes an intelligent early warning system and method for trunk road traffic safety based on edge computing to overcome the above-mentioned technical problems existing in the existing related technology.

[0019] To this end, the specific technical solution adopted by the present invention is as follows:

[0020] According to one aspect of the present invention, there is provided an intelligent early warning system for traffic safety of trunk roads based on edge computing, the system comprising: a power supply facility unit, a network facility unit, a traffic sensing facility unit, a traffic control and induction facility unit and an edge intelligent early warning device unit;

[0021] A power supply facility unit for providing power supply by combining power supply from the power grid and solar power supply;

[0022] Network facility unit, used to build communication network, realize data transmission and remote control;

[0023] A traffic sensing facility unit is used to sense the road traffic conditions of a road section using traffic sensing sensor equipment;

[0024] Traffic control and guidance facility unit, used to issue traffic control strategies using early warning prompts according to the different needs of the deployed sections, and provide warning information and guidance prompts to drivers of vehicles traveling on the corresponding sections;

[0025] The edge intelligent early warning device unit is used to monitor and warn the road traffic conditions of the road section in real time, realize the collection, processing, analysis and decision-making of road traffic information, and generate and issue early warning strategies.

[0026] Furthermore, the edge intelligent early warning device unit includes an edge computing module, a network switching module, and an operation and maintenance management module;

[0027] Edge computing module, used for automatic collection, processing, analysis and decision-making of road traffic information, and generation and issuance of early warning strategies;

[0028] Network exchange module, used to realize data communication, network connection and information interaction among various component modules;

[0029] The operation and maintenance management module is used to provide power guarantee for each component module, and to monitor the environmental status, power status, network status, and equipment operation status in real time to ensure the operation of the edge intelligent early warning device unit.

[0030] Furthermore, the edge computing module includes: a traffic information collection submodule, an intelligent analysis and decision-making submodule, and a traffic induction and warning submodule;

[0031] Traffic information collection submodule, used to collect road traffic information in real time using traffic sensing facility units;

[0032] Intelligent analysis and decision-making submodule, which is used to analyze the collected road traffic information based on edge computing, evaluate the road traffic status, identify traffic safety factors, and generate real-time warning strategies based on the identification results;

[0033] The traffic guidance and warning submodule is used to publish real-time warning strategies to the corresponding traffic control and guidance facility units.

[0034] Furthermore, the intelligent analysis and decision-making submodules include: traffic incident analysis and decision-making, road surface condition analysis and decision-making, and visibility analysis and decision-making;

[0035] Traffic incident analysis and decision-making, used to analyze the collected traffic detection video stream and generate early warning strategies based on the analysis results;

[0036] Pavement status analysis and decision-making, used to analyze the collected pavement data and generate early warning strategies based on the analysis results;

[0037] Visibility analysis and decision-making is used to analyze the collected image data and generate early warning strategies based on the analysis results.

[0038] Furthermore, analyzing the collected video data and generating an early warning strategy based on the analysis results includes the following steps:

[0039] Using cameras to collect traffic detection video streams;

[0040] Use the deep learning-based traffic incident detection algorithm to analyze video data and determine whether a traffic incident exists based on the analysis results;

[0041] If there is a traffic incident, a warning strategy is generated according to the type of traffic incident and the location where the traffic incident is detected; if there is no traffic incident, the traffic detection video stream continues to be read.

[0042] Furthermore, using a deep learning-based traffic incident detection algorithm to analyze video data and determining whether a traffic incident exists based on the analysis results includes the following steps:

[0043] Acquire traffic event image data and establish a traffic event image dataset;

[0044] Use the traffic incident image dataset to train the deep neural network, and obtain the trained deep neural network through repeated iterations;

[0045] Decode the collected traffic detection video stream to obtain continuous image data;

[0046] Input continuous image data into the trained deep neural network to obtain traffic event detection results.

[0047] Furthermore, analyzing the collected road surface data and generating an early warning strategy based on the analysis results includes the following steps:

[0048] Using a road surface state detector to collect road surface data to obtain road surface state data;

[0049] Analyze the road surface condition data and determine whether the road surface is icy based on the analysis results;

[0050] If there is ice on the road surface, a warning strategy is generated according to the location of the ice; if there is no ice on the road surface, the road surface data continues to be read.

[0051] Further, analyzing the collected image data and generating an early warning strategy based on the analysis results includes the following steps:

[0052] The visibility detector is used to collect image data to obtain the visibility of the road section;

[0053] The collected image data is analyzed using a foggy weather recognition algorithm based on dark channel priors, and foggy weather detection results are obtained based on the analysis results;

[0054] Based on the fog detection results and combined with the visibility of the road section, an early warning strategy is generated.

[0055] Furthermore, the collected image data is analyzed using a foggy weather recognition algorithm based on dark channel priors, and the foggy weather detection result is obtained based on the analysis result, including the following steps:

[0056] collecting image data using a visibility detector and decoding the image data;

[0057] The decoded image data is separated into channels, and the minimum value of each channel is selected as the grayscale value of the corresponding pixel to construct a grayscale image;

[0058] Count the grayscale values ​​of each pixel in the grayscale image and calculate the average value;

[0059] The average value is compared with the preset threshold value, and the fog detection result is obtained according to the comparison result.

[0060] According to another aspect of the present invention, there is also provided an intelligent early warning method for trunk road traffic safety based on edge computing, the method comprising the following steps:

[0061] S1. Use traffic sensing facilities to collect road traffic information in real time;

[0062] S2. Analyze the collected road traffic information based on edge computing, evaluate the road traffic status, identify traffic safety factors, and generate real-time warning strategies based on the identification results;

[0063] S3. Publish the real-time warning strategy to the corresponding traffic control and guidance facilities.

[0064] The beneficial effects of the present invention are:

[0065] 1. By deploying edge computing technology along trunk roads, the present invention can directly process and analyze data near the data source, reduce data transmission delays, realize real-time monitoring of traffic conditions and rapid identification of anomalies, effectively shorten the time from detection to response, and greatly improve the efficiency and accuracy of traffic safety emergency response.

[0066] 2. Through distributed deployment and edge computing capabilities, the present invention can feed back road information to vehicles in the first place, understand accurate road conditions in advance, avoid unnecessary delays and congestion, and improve road operation efficiency and traffic capacity.

[0067] 3. The present invention uses edge computing technology to create an intelligent traffic safety warning system, which can realize real-time monitoring and data analysis of the road conditions of ordinary national and provincial trunk roads. The precise road condition information provided by the system can realize more intelligent traffic management and control, and provide accurate travel suggestions, thereby improving the quality and efficiency of road operation services and comprehensively improving the road traffic operation experience.

[0068] 4. The present invention realizes accurate monitoring and early warning of key areas by deploying distributed edge intelligent early warning subsystems in key sections. This targeted deployment strategy can not only reduce the construction and operation costs of the system, but also improve the safety of the road sections. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0070] Figure 1 is a principle block diagram of an intelligent early warning system for trunk road traffic safety based on edge computing according to an embodiment of the present invention;

[0071] Figure 2 is a general architecture diagram of an intelligent early warning system for trunk road traffic safety based on edge computing according to an embodiment of the present invention;

[0072] Figure 3 It is a composition structure diagram of the edge intelligent warning subsystem in the trunk road traffic safety intelligent warning system based on edge computing according to an embodiment of the present invention;

[0073] Figure 4 It is a composition structure diagram of an edge intelligent warning device unit in an arterial highway traffic safety intelligent warning system based on edge computing according to an embodiment of the present invention;

[0074] Figure 5is a flow chart of a traffic incident detection algorithm based on deep learning in an intelligent early warning system for trunk road traffic safety based on edge computing according to an embodiment of the present invention;

[0075] Figure 6 It is a flow chart of a foggy day recognition algorithm based on dark channel data statistics in an intelligent early warning system for trunk road traffic safety based on edge computing according to an embodiment of the present invention;

[0076] Figure 7 is a schematic diagram of an embodiment of an intelligent early warning system for traffic safety of trunk roads based on edge computing according to an embodiment of the present invention;

[0077] Figure 8 It is a composition diagram of an edge intelligent early warning subsystem for accident-prone sections in an intelligent early warning system for traffic safety of trunk roads based on edge computing according to an embodiment of the present invention;

[0078] Fig. 9 It is a flow chart of edge intelligent warning for accident-prone sections in an intelligent warning system for traffic safety of trunk roads based on edge computing according to an embodiment of the present invention;

[0079] Fig.10 It is a composition diagram of an edge intelligent warning subsystem for icy road sections in an intelligent warning system for trunk road traffic safety based on edge computing according to an embodiment of the present invention;

[0080] Fig.11 It is a flow chart of edge intelligent warning for icy road sections in an intelligent warning system for traffic safety of trunk roads based on edge computing according to an embodiment of the present invention;

[0081] Fig.12 It is a composition diagram of an edge intelligent warning subsystem for fog-prone sections in an intelligent warning system for trunk road traffic safety based on edge computing according to an embodiment of the present invention;

[0082] Fig.13 It is a flow chart of edge intelligent warning for fog-prone sections in an intelligent warning system for traffic safety of trunk roads based on edge computing according to an embodiment of the present invention;

[0083] Fig.14 is a flow chart of an intelligent early warning method for trunk road traffic safety based on edge computing according to an embodiment of the present invention;

[0084] Fig.15 It is a flow chart of an intelligent early warning method for trunk road traffic safety based on edge computing according to an embodiment of the present invention;

[0085] Fig.16 It is the overall architecture diagram of the video centralized monitoring technology in the prior art.

[0086] In the figure:

[0087] 1. Power supply facility unit; 2. Network facility unit; 3. Traffic perception facility unit; 4. Traffic control and induction facility unit; 5. Edge intelligent early warning device unit. DETAILED DESCRIPTION

[0088] To further illustrate each embodiment, the present invention provides drawings, which are part of the disclosure of the present invention and are mainly used to illustrate the embodiments. They can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these contents, ordinary technicians in this field should be able to understand other possible implementation methods and advantages of the present invention.

[0089] According to an embodiment of the present invention, an intelligent early warning system and method for trunk road traffic safety based on edge computing are provided.

[0090] The present invention is further described with reference to the accompanying drawings and specific embodiments. Figure 1 , Figure 3-Figure 7 As shown, according to an embodiment of the present invention, an intelligent early warning system for trunk road traffic safety based on edge computing includes: a power supply facility unit 1, a network facility unit 2, a traffic sensing facility unit 3, a traffic control and induction facility unit 4, and an edge intelligent early warning device unit 5;

[0091] The power supply facility unit 1 is used to provide power supply by combining power supply from the power grid and solar power supply.

[0092] It should be noted that, according to the infrastructure conditions of the highway, the edge intelligent early warning device unit 5 is powered by a combination of power grid power supply and solar power supply.

[0093] The network facility unit 2 is used to build a communication network to achieve data transmission and remote control.

[0094] It should be noted that, according to the infrastructure conditions of the highway, a stable communication network is constructed through wired or wireless means to realize data transmission and remote control between the edge intelligent early warning device unit 5 and the management center.

[0095] The traffic sensing facility unit 3 is used to sense the road traffic conditions of the road section using traffic sensing sensor equipment.

[0096] It should be noted that, according to the monitoring requirements of the deployed road sections, traffic perception sensor equipment such as cameras, radars, road surface condition detectors, and visibility detectors are used to perceive the road traffic conditions of the road sections.

[0097] The traffic control and guidance facility unit 4 is used to issue traffic control strategies using early warning prompts according to different needs of the deployed sections, and provide warning information and guidance prompts to vehicle drivers traveling on the corresponding sections.

[0098] It should be noted that, according to the different needs of the deployed sections, the traffic control strategy generated by the edge intelligent early warning device unit 5 is released using early warning prompts such as variable information boards, sound and light alarms, and yellow flashing warning lights to provide warning information and induction prompts to vehicle drivers traveling on the section.

[0099] The edge intelligent early warning device unit 5 is used to monitor and warn the road traffic conditions of the road section in real time, realize the collection, processing, analysis and decision-making of road traffic information, and generate and issue early warning strategies.

[0100] It should be noted that the road traffic conditions of the road section are monitored and warned in real time, the road traffic information is collected, processed, analyzed and decided, a local warning strategy is generated, and the warning strategy is published to the corresponding traffic control and guidance facility unit 4.

[0101] Among them, the software of the edge intelligent early warning device unit 5 adopts the Linux operating system and modular functional design, and is implemented by GO language programming.

[0102] Specifically, the edge intelligent early warning device unit 5 includes an edge computing module, a network switching module, and an operation and maintenance management module;

[0103] The edge computing module is used for automatic collection, processing, analysis and decision-making of road traffic information, and for generating and issuing early warning strategies.

[0104] It should be noted that the edge computing module consists of an edge computing industrial computer as the core computing unit of the edge intelligent early warning device unit 5, which automatically collects, processes, analyzes and makes decisions on road traffic information, generates early warning strategies, and automatically publishes the strategies to the corresponding traffic control and induction facility unit 4.

[0105] Among them, traffic information is collected through the edge intelligent warning device unit 5, and road traffic information (such as traffic events, road surface conditions, visibility, etc.) detected by traffic sensing facilities (such as cameras, road surface condition detectors, visibility detectors, etc.) is collected in real time; based on edge computing, the collected road traffic information is intelligently analyzed, the current road traffic status is evaluated, and the factors affecting traffic safety, such as traffic congestion, accidents, low visibility, road icing, etc., are analyzed, and local traffic condition warnings, road structure warnings, safety warning controls and other real-time warning strategies are generated.

[0106] In addition, the edge computing module uses NVIDIA Jetson Xavier NX edge computing industrial computer as the core computing unit, which is an ARMv8 processor architecture with 21TOPS AI computing power.

[0107] Specifically, the edge computing module includes: traffic information collection submodule, intelligent analysis and decision-making submodule, and traffic guidance and warning submodule;

[0108] The traffic information collection submodule is used to collect road traffic information in real time using traffic sensing facility units.

[0109] It should be supplemented that the road traffic information detected by the traffic sensing facility unit 3 is collected in real time, including a camera, a road surface state detector and a visibility detector.

[0110] The intelligent analysis and decision-making submodule is used to analyze the collected road traffic information based on edge computing, evaluate the road traffic status, identify traffic safety factors, and generate real-time warning strategies based on the identification results.

[0111] It should be noted that the collected road traffic information is intelligently analyzed to evaluate the current road traffic status, analyze the factors affecting traffic safety, and generate local early warning strategies, including traffic event detection and early warning prompts, road icing detection and early warning prompts, and fog detection and early warning control.

[0112] Specifically, the intelligent analysis and decision-making submodules include: traffic incident analysis and decision-making, road surface condition analysis and decision-making, and visibility analysis and decision-making;

[0113] Traffic incident analysis and decision-making is used to analyze the collected traffic detection video stream and generate early warning strategies based on the analysis results.

[0114] Specifically, analyzing the collected video data and generating an early warning strategy based on the analysis results includes the following steps:

[0115] Use cameras to collect traffic detection video streams.

[0116] Use the deep learning-based traffic incident detection algorithm to analyze the video data, and determine whether a traffic incident occurs based on the analysis results.

[0117] It should be noted that traffic incidents include abnormal parking, speeding, vehicles driving in the wrong direction, traffic congestion, traffic accidents, and spilled objects.

[0118] Specifically, using a deep learning-based traffic incident detection algorithm to analyze video data and judging whether a traffic incident exists based on the analysis results includes the following steps:

[0119] Acquire traffic event image data and establish a traffic event image dataset;

[0120] Use the traffic incident image dataset to train the deep neural network, and obtain the trained deep neural network through repeated iterations;

[0121] Decode the collected traffic detection video stream to obtain continuous image data;

[0122] Input continuous image data into the trained deep neural network to obtain traffic event detection results.

[0123] If there is a traffic incident, a warning strategy is generated according to the type of traffic incident and the location where the traffic incident is detected; if there is no traffic incident, the traffic detection video stream continues to be read.

[0124] It should be noted that, for example, a warning message is issued through a variable information board 1 km upstream of the event location: "A traffic accident has occurred 1 km ahead, please slow down"; at the same time, the warning information and on-site pictures are reported to the central auxiliary subsystem.

[0125] Pavement condition analysis and decision-making are used to analyze the collected pavement data and generate early warning strategies based on the analysis results.

[0126] Specifically, analyzing the collected road surface data and generating an early warning strategy based on the analysis results includes the following steps:

[0127] Using a road surface state detector to collect road surface data to obtain road surface state data;

[0128] Analyze the road surface condition data and determine whether the road surface is icy based on the analysis results;

[0129] If there is ice on the road surface, a warning strategy is generated according to the location of the ice; if there is no ice on the road surface, the road surface data continues to be read.

[0130] It should be noted that the road condition data includes road icing, snow accumulation, water accumulation, road temperature and other condition data; for example, a warning message is issued through a variable information board 1 km upstream of the icing location: "There is road icing 1 km ahead, please slow down"; at the same time, the warning information is reported to the central auxiliary subsystem, and if a camera is installed on site, the on-site picture will be reported together.

[0131] Visibility analysis and decision-making is used to analyze the collected image data and generate early warning strategies based on the analysis results.

[0132] Specifically, analyzing the collected image data and generating an early warning strategy based on the analysis results includes the following steps:

[0133] The visibility detector is used to collect image data to obtain the visibility of the road section.

[0134] The collected image data is analyzed using a foggy day recognition algorithm based on dark channel prior, and the foggy day detection results are obtained based on the analysis results.

[0135] It should be noted that the dark channel prior is a basic assumption that in most fog-free image areas, at least one pixel has a channel value (R, G, B color channels) close to zero.

[0136] Specifically, using the fog recognition algorithm based on dark channel prior to analyze the collected image data, and obtaining the fog detection result based on the analysis result includes the following steps:

[0137] collecting image data using a visibility detector and decoding the image data;

[0138] The decoded image data is separated into channels, and the minimum value of each channel is selected as the grayscale value of the corresponding pixel to construct a grayscale image;

[0139] Count the grayscale values ​​of each pixel in the grayscale image and calculate the average value;

[0140] The average value is compared with the preset threshold value, and the fog detection result is obtained according to the comparison result.

[0141] It should be noted that the preset threshold is the fog data threshold. The fog data threshold is set by collecting foggy images on the tested section, dividing the area according to the characteristics of the image, dividing the sky area and the non-sky area, and performing dark channel calculation on the non-sky area. The specific calculation method is to take the minimum value in the three RGB channels of the image area to form a grayscale image, and then count the average grayscale value of each pixel in the grayscale image, and set this value as the fog data threshold.

[0142] In addition, if the average value is greater than or equal to the preset threshold, it is considered that there is fog; if the average value is less than the preset threshold, it is considered that there is no fog.

[0143] Based on the fog detection results and combined with the visibility of the road section, an early warning strategy is generated.

[0144] It should be noted that if the fog detection result is foggy and the visibility is lower than the set visibility threshold, an early warning strategy will be generated, such as turning on the fog-inducing yellow flashing light. At the same time, the early warning information and on-site pictures will be reported to the central auxiliary subsystem; if the fog detection result is no fog, the image data will continue to be read.

[0145] The traffic guidance and warning submodule is used to publish the real-time warning strategy to the corresponding traffic control and guidance facility unit 4.

[0146] It should be noted that the warning strategy is published to the corresponding traffic control and guidance facility unit 4 (including publishing the warning prompt information to the variable information board, turning on the fog guidance yellow flashing light, etc.) through the edge intelligent warning device unit 5, so as to realize the autonomous publication and control of traffic guidance and warning.

[0147] The network switching module is used to realize data communication, network connection and information interaction among various component modules.

[0148] It should be noted that the network switching module includes components such as an industrial Ethernet switch and a fiber optic terminal box, which are used to realize data communication between the various components inside the edge intelligent early warning device unit 5, and to connect with external network facilities to realize information interaction between the edge intelligent early warning device unit 5 and the management center.

[0149] Among them, the network switching module uses an industrial Ethernet switch as the network switching unit, providing 2 Gigabit fiber optic interfaces and 8 Gigabit RJ45 network ports.

[0150] The operation and maintenance management module is used to provide power guarantee for each component module, and to monitor the environmental status, power status, network status, and equipment operation status in real time to ensure the operation of the edge intelligent early warning device unit.

[0151] It should be noted that the operation and maintenance management module includes a chassis, an operation and maintenance management controller, and operation and maintenance management sensors (including automatic reclosing, environmental monitoring sensors, power supply lightning arresters, power supply output ports) and other components, which are connected to external power supply facilities to provide power guarantee for each component module inside the edge intelligent early warning device unit 5. At the same time, it monitors the environmental status, power status, network status, equipment operation status, etc. in real time to ensure the normal operation of the edge intelligent early warning device.

[0152] Among them, the operation and maintenance management module adopts an integrated chassis form, which includes a chassis body, controller, automatic reclosing switch, LED lighting, temperature-controlled fan, power lightning arrester, power output port and other components to provide a power operating environment for the edge computing module and network switching module.

[0153] In addition, if Figure 2 As shown in the figure, each edge intelligent warning subsystem, supported by edge computing, has the capabilities of autonomous perception, autonomous analysis, autonomous decision-making and autonomous control, and can monitor and warn of road traffic conditions on each road section in real time without relying on the center. At the same time, each edge intelligent warning subsystem can also synchronously report the monitoring and warning information of each road section to the central auxiliary subsystem, and receive and execute the road network overall management commands issued by the central system.

[0154] In one embodiment, Figure 8 As shown, for accident-prone sections, the edge intelligent warning subsystem is deployed at accident-prone sections such as level intersections, U-turns, continuous curves, and long downhills on ordinary national and provincial trunk roads to conduct real-time automatic detection and warning of the traffic operation status and traffic incident conditions on accident-prone sections. The system includes an edge intelligent warning device, a camera, and a variable information board.

[0155] Through video surveillance of accident-prone sections of the road through cameras, intelligent analysis of surveillance videos is carried out through edge intelligent devices, and real-time edge intelligent detection is carried out on various types of traffic incidents such as abnormal parking, traffic congestion, vehicle reversing, traffic accidents, and spilled objects. Once a traffic incident is detected, local warning information is generated and released through the variable information board. At the same time, the warning information and on-site pictures are reported to the central auxiliary subsystem through the 4G / 5G network.

[0156] like Fig. 9 As shown, for the edge intelligent early warning method of accident-prone road sections, the method includes:

[0157] Step 1: Collect surveillance video from the camera;

[0158] Step 2: Perform intelligent analysis on the surveillance video and generate warning information if abnormal parking, traffic congestion, vehicle reversal, traffic accident, spilled objects and other traffic events are detected;

[0159] Step 3: Publish the warning information to the variable information board, and at the same time, report the warning information and on-site pictures to the central auxiliary subsystem.

[0160] In one embodiment, Fig.10 As shown, the edge intelligent warning subsystem for icy road sections is deployed on common icy road sections of ordinary national and provincial trunk roads to perform real-time detection and warning of the icy road conditions on the sections. The system includes an edge intelligent warning device, a camera, a road state detector and a variable information board.

[0161] The road surface condition detector monitors the road surface conditions such as ice, snow accumulation, water accumulation, and road surface temperature in real time. The collected road surface conditions are intelligently analyzed through edge intelligent devices to evaluate the icing conditions on the road surface. Once the risk of icing is detected, local warning information is generated and released through the variable information board. At the same time, the warning information and on-site pictures are reported to the central auxiliary subsystem through the 4G / 5G network.

[0162] like Fig.11 As shown, for an intelligent early warning method for an icy road section edge, the method includes:

[0163] Step 1: Collect road conditions such as road icing, snow accumulation, water accumulation, road temperature, etc. detected by the road condition detector, as well as the video detected by the camera;

[0164] Step 2: Intelligently analyze the collected road conditions to evaluate the icing conditions on the road. If icing risk is detected, generate warning information;

[0165] Step 3: Publish the warning information to the variable information board, and at the same time, report the warning information and on-site pictures to the central auxiliary subsystem.

[0166] In one embodiment, Fig.12 As shown, the edge intelligent warning subsystem for sections prone to fog is deployed on ordinary national and provincial trunk roads in sections prone to fog to conduct real-time detection and warning of visibility on the section. The system includes an edge intelligent warning device, a camera, a visibility detector and a yellow flashing warning light.

[0167] The visibility of the road section is monitored in real time through the visibility detector, and video surveillance is carried out through the camera. The edge intelligent device analyzes the collected visibility information. At the same time, the image recognition algorithm is used to intelligently analyze the on-site video. The weather conditions of the road section are evaluated based on the comprehensive analysis results. Once the risk of fog is detected, a local early warning strategy is generated to control the yellow flashing lights deployed along the road to turn on and flash, and guide the path of the moving vehicles; at the same time, the early warning information and on-site pictures are reported to the central auxiliary subsystem through the 4G / 5G network.

[0168] like Fig.13 As shown, for the edge intelligent early warning method of a road section prone to fog, the method includes:

[0169] Step 1: Collect visibility information detected by the visibility detector and video detected by the camera;

[0170] Step 2: Analyze the collected visibility information and intelligently analyze the collected surveillance videos. Evaluate the weather conditions of the road section based on the comprehensive analysis results. If the risk of fog is detected, generate an early warning control strategy.

[0171] Step 3: According to the early warning control strategy, the yellow flashing light is controlled to turn on and flash, and the early warning information and on-site video are reported to the central auxiliary subsystem.

[0172] like Figure 14-15 As shown, according to another embodiment of the present invention, a trunk road traffic safety intelligent warning method based on edge computing is also provided, and the method includes the following steps:

[0173] S1. Use traffic sensing facilities to collect road traffic information in real time;

[0174] S2. Analyze the collected road traffic information based on edge computing, evaluate the road traffic status, identify traffic safety factors, and generate real-time warning strategies based on the identification results;

[0175] S3. Publish the real-time warning strategy to the corresponding traffic control and guidance facilities.

[0176] In summary, with the help of the above technical solution of the present invention, by deploying edge computing technology along the trunk roads, data processing and analysis can be performed directly near the data source, reducing data transmission delays, realizing real-time monitoring of traffic conditions and rapid identification of abnormalities, effectively shortening the time from detection to response, and greatly improving the efficiency and accuracy of traffic safety emergency handling. Through distributed deployment and edge computing capabilities, road information can be fed back to vehicles in the first time, accurate road conditions can be understood in advance, unnecessary delays and congestion can be avoided, and road operation efficiency and traffic capacity can be improved. By creating an intelligent traffic safety warning system through edge computing technology, real-time monitoring and data analysis of the road conditions of ordinary national and provincial trunk roads can be achieved. Through the accurate road condition information provided by the system, more intelligent traffic management and control can be achieved, and accurate travel suggestions can be provided, thereby improving the quality and efficiency of road operation services and comprehensively improving the road traffic operation experience. By deploying a distributed edge intelligent warning subsystem in key sections, accurate monitoring and warning of key areas are achieved. This targeted deployment strategy can not only reduce the construction and operation costs of the system, but also improve the safety of the road section.

[0177] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. The intelligent early warning system for trunk road traffic safety based on edge computing is characterized by: The system includes: a power supply facility unit, a network facility unit, a traffic sensing facility unit, a traffic control and induction facility unit, and an edge intelligent early warning device unit; The power supply facility unit is used to provide power supply by combining power supply from the power network and power supply from solar energy; The network facility unit is used to build a communication network to achieve data transmission and remote control; The traffic sensing facility unit is used to sense the road traffic conditions of the road section using traffic sensing sensor equipment; The traffic control and guidance facility unit is used to issue traffic control strategies using early warning prompts according to different requirements of the deployed road sections, and provide warning information and guidance prompts to vehicle drivers traveling on the corresponding road sections; The edge intelligent early warning device unit is used to monitor and warn the road traffic conditions of the road section in real time, realize the collection, processing, analysis and decision-making of road traffic information, and generate and issue early warning strategies; The edge intelligent early warning device unit includes an edge computing module, a network switching module, and an operation and maintenance management module; The edge computing module is used for automatic collection, processing, analysis and decision-making of road traffic information, and generation and issuance of early warning strategies; The network exchange module is used to realize data communication, network connection and information interaction among various component modules; The operation and maintenance management module is used to provide power guarantee for each component module, and to monitor the environmental status, power status, network status, and equipment operation status in real time to ensure the operation of the edge intelligent early warning device unit; The edge computing module includes: a traffic information collection submodule, an intelligent analysis and decision-making submodule, and a traffic guidance and warning submodule; The traffic information collection submodule is used to collect road traffic information in real time using the traffic sensing facility unit; The intelligent analysis and decision-making submodule is used to analyze the collected road traffic information based on edge computing, evaluate the road traffic status, identify traffic safety factors, and generate real-time warning strategies based on the identification results; The traffic guidance and warning submodule is used to publish the real-time warning strategy to the corresponding traffic control and guidance facility units; The intelligent analysis and decision submodule includes: traffic incident analysis and decision, road surface condition analysis and decision, and visibility analysis and decision; The traffic event analysis and decision-making is used to analyze the collected traffic detection video stream and generate an early warning strategy based on the analysis results; The road surface state analysis and decision-making is used to analyze the collected road surface data and generate an early warning strategy based on the analysis results; the analysis of the collected road surface data and the generation of an early warning strategy based on the analysis results include the following steps: The pavement status detector is used to collect pavement data to obtain pavement status data, wherein the pavement status data includes pavement icing status data, snow accumulation status data, water accumulation status data and pavement temperature status data; Analyze the road surface condition data and determine whether the road surface is icy based on the analysis results; If the road surface is icy, a warning strategy is generated based on the location of the ice; if the road surface is not icy, the road surface data continues to be read; The visibility analysis and decision-making is used to analyze the collected image data and generate an early warning strategy according to the analysis results; the analysis of the collected image data and generating an early warning strategy according to the analysis results includes the following steps: The visibility detector is used to collect image data to obtain the visibility of the road section; The collected image data is analyzed using a foggy weather recognition algorithm based on dark channel priors, and foggy weather detection results are obtained based on the analysis results, including the following steps: collecting image data using a visibility detector and decoding the image data; The decoded image data is separated into channels, and the minimum value of each channel is selected as the grayscale value of the corresponding pixel to construct a grayscale image; Count the grayscale values ​​of each pixel in the grayscale image and calculate the average value; The average value is compared with a preset threshold value, and a fog detection result is obtained according to the comparison result; Among them, the preset threshold is the fog data threshold. The fog data threshold is set by collecting foggy images on the tested road section, dividing the area according to the characteristics of the image, dividing the sky area and the non-sky area, and performing dark channel calculation on the non-sky area. The specific calculation method is to take the minimum value of the three RGB channels in the image area to form a grayscale image, and then count the average grayscale value of each pixel in the grayscale image, and set the value as the fog data threshold; If the average value is greater than or equal to the preset threshold, it is considered that there is fog; if the average value is less than the preset threshold, it is considered that there is no fog; Generate early warning strategies based on fog detection results and visibility of road sections; Among them, if the fog detection result is foggy and the visibility is lower than the set visibility threshold, an early warning strategy is generated, the fog-inducing yellow flashing light is turned on, and the early warning information and on-site pictures are reported to the central auxiliary subsystem; if the fog detection result is no fog, the image data continues to be read.

2. According to claim 1, the intelligent early warning system for trunk road traffic safety based on edge computing is characterized in that: The step of analyzing the collected video data and generating an early warning strategy based on the analysis results includes the following steps: Using cameras to collect traffic detection video streams; Use the deep learning-based traffic incident detection algorithm to analyze video data and determine whether a traffic incident exists based on the analysis results; If there is a traffic incident, a warning strategy is generated according to the type of traffic incident and the location where the traffic incident is detected; if there is no traffic incident, the traffic detection video stream continues to be read.

3. The intelligent early warning system for trunk road traffic safety based on edge computing according to claim 2 is characterized in that: The method of analyzing the video data using a traffic incident detection algorithm based on deep learning and determining whether a traffic incident exists according to the analysis results includes the following steps: Acquire traffic event image data and establish a traffic event image dataset; Use the traffic incident image dataset to train the deep neural network, and obtain the trained deep neural network through repeated iterations; Decode the collected traffic detection video stream to obtain continuous image data; Input continuous image data into the trained deep neural network to obtain traffic event detection results.

4. An intelligent early warning method for trunk road traffic safety based on edge computing, used to implement an intelligent early warning system for trunk road traffic safety based on edge computing as claimed in any one of claims 1 to 3, characterized in that: The method comprises the following steps: S1. Use traffic sensing facilities to collect road traffic information in real time; S2. Analyze the collected road traffic information based on edge computing, evaluate the road traffic status, identify traffic safety factors, and generate real-time warning strategies based on the identification results; S3. Publish the real-time warning strategy to the corresponding traffic control and guidance facilities.