Road traffic potential safety hazard troubleshooting method based on tunnel road section scene

Abstract

The invention discloses a road traffic potential safety hazard troubleshooting method based on a tunnel road section scene. The method comprises the following steps: S1, constructing a tunnel road section road traffic potential safety hazard troubleshooting knowledge system; s2, establishing a standardized data acquisition process; s3, carrying out cleaning based on the road section data of the map; s4, systematized stake number and troubleshooting road section calibration is carried out; s5, establishing intelligent tunnel road section identification; s6, a tunnel road section road traffic potential safety hazard standardized troubleshooting method and a conclusion are obtained; the AI recognition technology, the laser radar and the intelligent vehicle-mounted equipment are combined, and automatic collection and troubleshooting of traffic signs, marking lines and safety facilities of tunnel road sections are achieved. Through AI automatic kilometer pile identification and manual auxiliary confirmation, the starting and ending points and the up-down direction of the road section are accurately calibrated, the uniqueness of a troubleshooting result is ensured through a data cleaning technology, the troubleshooting efficiency and accuracy of traffic hidden dangers are remarkably improved, and the traffic safety management level of the tunnel road section is effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of road traffic safety, and in particular to a method for detecting road traffic safety hazards based on a tunnel section scenario. Background Art

[0002] In recent years, road infrastructure construction has been vigorously carried out. In actual road construction, considering the limitations of safety, economy, humanities and ecological environment, natural geography, and hydrogeology, it is inevitable to build highway tunnels that cross mountains to reduce driving distance. With the substantial increase in highway mileage, the accompanying traffic safety issues are becoming increasingly severe. Tunnels and other sections are also prone to traffic accidents, causing a large number of casualties and economic losses. Therefore, studying the method of troubleshooting tunnel traffic hazards is of great significance to improving traffic safety in tunnels.

[0003] The existing technology has the following deficiencies: there is no specific identification method for the current inspection of road traffic safety hazards in tunnel sections, and there is a lack of means to focus on whether there are safety hazards in the entire section, and there is a lack of theoretical support. At the same time, no detailed analysis of the detailed hazard points inside the tunnel section has been conducted. The inspection results are difficult to effectively promote relevant departments to carry out road traffic safety hazard inspection work in tunnel sections.

[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0005] The purpose of the present invention is to provide a method for detecting road traffic safety hazards based on a tunnel section scenario to solve the problems in the above-mentioned background technology.

[0006] In order to achieve the above object, the present invention provides the following technical solution: comprising the following steps: S1. Construct a knowledge system for checking road traffic safety hazards in tunnel sections; S2. Establish a standardized data collection process; S3, cleaning the road section data based on the map; S4, perform systematic stake number and inspection section calibration; S5. Establish intelligent tunnel section identification; S6. Standardized inspection method and conclusion of road traffic safety hazards in tunnel sections; In S1, the specific steps are as follows: 1.1. Collect basic parameters and structural information of the tunnel section.

[0007] 1.2 Traffic signs and road markings: The system identifies and checks traffic signs and road markings inside and outside the tunnel, including: Traffic markings: Check the dividing lines of the up and down lanes, the edge lines of the lanes, and the vertical markings.

[0008] Traffic signs: Check upstream and downstream tunnel signs, headlight signs, height limit signs, width limit signs, speed limit signs, no overtaking signs, fire equipment signs, emergency telephone signs, evacuation signs, tunnel exit distance warning signs, emergency parking lane signs, and speed limit removal signs.

[0009] 1.3. Safety facilities: Check raised road signs, contour markers, tunnel contour strips, emergency parking strips, escape routes, emergency telephones, and fire-fighting facilities.

[0010] 1.4. Technological facilities: testing lane signal lights and monitoring equipment technological facilities.

[0011] 1.5. Other facilities: Check the lighting facilities and height limit frames inside and outside the tunnel.

[0012] In S2, the specific steps are as follows: Equipment layout: Utilize multiple high-definition cameras installed on the modified vehicle to collect road image data at high frequency from the left front, front and right front perspectives; GPS information collection: Use the GPS measuring instrument to obtain real-time position data of the vehicle during driving; Vehicle status monitoring: Use the inertial navigation system to collect dynamic information of roll angle, pitch angle and yaw angle, record the spatial posture of the vehicle, combine the speed sensor to collect vehicle speed information, and use the altitude measuring instrument to obtain the altitude of the vehicle; Obstacle detection: Use laser radar to perform real-time detection of the front, left and right sides of the vehicle, and record the existence and distance information of obstacles; Data storage and synchronization: All collected images, GPS, vehicle speed and altitude information are synchronized through time tags, stored in their respective equipment storage centers, and matched and used during the troubleshooting process.

[0013] Preferably, in S3, the specific steps are as follows: Data digitization processing: Using the video, GPS location, time and other information automatically collected by the vehicle-side equipment, the traffic signs and markings in the tunnel are digitized to generate corresponding image data; Data mapping: Upload digital road signs and road marking information to Baidu Maps and other map platforms based on their GPS data to create corresponding road atlases; Manual cleaning: For duplicate data generated during the collection process due to reasons such as waiting for red lights, manual cleaning is used to eliminate redundant or irrelevant image data to ensure that only the road section data that needs to be checked is retained.

[0014] Preferably, in S4, the specific steps are as follows: 4.1. AI automatic identification: AI technology is used to identify kilometer posts and hundred-meter posts on the road, and each picture is matched with its corresponding post number to generate an electronic post number; Direction calibration: The up and down directions of the tunnel section are calibrated by kilometer posts, and the images collected by the main camera and the right camera are used for auxiliary calibration; Manual confirmation: Based on AI identification, the accuracy of the kilometer post calibration is further confirmed manually to ensure that the calibration result is consistent with the actual situation; Using the image of the right camera as a reference, manual button calibration is performed on the image with kilometer post marks to generate the corresponding electronic post number; 4.2. Task creation: According to the requirements of the inspection task, determine the road section information that needs to be inspected; Determine the road section information: According to the collected image information, determine the basic data such as the name, area, jurisdiction, uplink and downlink directions, starting and ending point numbers of each road section to be inspected; Calibration content: Complete the calibration of key information such as the road section name, inspection mileage, data collection time, speed limit, etc.

[0015] Preferably, in S5, the specific steps are as follows: when conducting the inspection of road traffic safety hazards in tunnel sections, it is necessary to realize intelligent identification of tunnel sections before the inspection of tunnel sections, determine the corresponding starting and ending point position information of the tunnel sections, and combine the intelligent identification with the front-end perception and manual assistance means for dual identification; perform obstacle discrimination on the signal data returned by the front-end perception device, identify and calculate the basic parameters of the section, dynamically identify and perceive the tunnel section for the slight or differential changes in the cross-sectional width of the road in different tunnel sections, and use the tunnel section perception discrimination algorithm to calculate and manually assist the identification of the tunnel section. The tunnel section discrimination method is as follows: the laser radar emits a laser beam at a certain frequency to detect whether there are walls, metals and other objects that are characteristic of the tunnel section directly above, on the left and on the right of the vehicle; compare the received signal reflected from the target with the transmitted signal to create a set of point clouds; measure the distance to the wall by calculating the time difference between emission and reflection; the tunnel section recognition algorithm is designed as follows: Step 1: Automatic calculation time The reflection distance of the radar receiver directly above the point is , where the reflection distance calculation formula is , c represents the propagation speed of laser in air, T represents the time interval from light beam emission to reception; When there is no reflection, T is 0 Step2: Judgment Is between 2.5 and 15: like <2.5 or >15, the point does not belong to the tunnel section; If 2.5≤ ≤15, go to Step 3; Step 3: Calculation time Point right radar receiver reflection distance ; Step4: Judgment Is it greater than 0.5? like ≤0.5, the point does not belong to the tunnel section; like >0.5, then go to Step 5; Step 5: Calculation time Point left radar receiver reflection distance ; Step 6: Judgement Is it greater than 0.5? like ≤0.5, the point does not belong to the tunnel section; like >0.5, the point may be in a tunnel section, record the starting point Point is the starting point of the tunnel section , go to Step 7; Step 7: Order ,calculate time The reflection distance of the radar receiver directly above the point is ,in is the data collection time interval; Step 8: Judgement Is between 2.5 and 15: like <2.5 or >15, the point does not belong to the tunnel section; If 2.5≤ ≤15, go to Step 9; Step 9: Calculation time Point right radar receiver reflection distance ; Step10: Judgment Is it greater than 0.5? like ≤0.5, the point does not belong to the tunnel section; like >0.5, then go to Step 11; Step 11: Calculation time Point left radar receiver reflection distance ; Step12: Judgment Is it greater than 0.5? like ≤0.5, the point does not belong to the tunnel section; like >0.5, the point may be in a tunnel section. Point and tunnel section start Distance , go to Step 13; Step13: Judgment Is it greater than 42m? like <42m, then the road section may be under a bridge; like ≥42m, the road section is in a tunnel section; Step 14: Construct a set of points in the tunnel section in chronological order ,definition and The starting point and end point of the tunnel section are shown respectively, and then go to Step 15; Step15: Judgment Is it greater than 3000m? like <3000m, the tunnel does not belong to the extra-long tunnel section, and the process goes to Step 16; like ≥3000m, the tunnel is a special long tunnel section; Step16: Judgment Is it greater than 1000m? like <1000m, the tunnel does not belong to the long tunnel section, and go to Step 17; like ≥1000m, the tunnel is a long tunnel section; Step17: Judgment Is it greater than 500m? like <500m, the tunnel is a short tunnel section; like ≥500m, the tunnel belongs to the middle tunnel section; The tunnel section data discrimination algorithm is used to identify tunnel sections with not limited to constant cross-sectional width and dynamically changing cross-sectional width, determine the starting and ending points, and identify and judge the tunnel length category.

[0016] Preferably, in S6, the specific steps are as follows: the tunnel section road traffic safety hazard inspection method is based on the constructed safety hazard inspection knowledge system, and the tunnel section is inspected from different aspects of traffic signs, traffic markings, safety facilities, scientific and technological facilities and other facilities; firstly, the layout positions of the tunnel section traffic signs, traffic markings, safety facilities and the like are inspected through the speed limit of the section in the basic information, and after the basic information database of the section is constructed, the contents of the tunnel section traffic signs, traffic markings, safety facilities, scientific and technological facilities and other facilities can be clarified; the tunnel section inspection method relies on intelligently collected pictures to visually inspect the various hazard contents of the tunnel section; the inspection steps are as follows: Step 1: Determine the starting and ending points of the tunnel section ; Step 2: According to the standard provisions in the knowledge system for safety hazard inspection and the location information of safety hazards in the tunnel section, determine the best points for inspection of traffic signs, traffic markings, safety facilities, scientific and technological facilities and other facilities. ;in , The distance or range between the inspection points and the reference points for traffic signs, traffic markings, safety facilities, scientific and technological facilities and other facilities specified in relevant standards; Step 3: Determine the best point The picture below is the best angle for troubleshooting safety hazards; Step 4: Confirm whether traffic signs, traffic markings, safety facilities, scientific and technological facilities and other facilities are missing or whether there are safety hazards based on the image from the best viewing angle: If there are no safety hazards at the best viewing angle, the hidden danger inspection for this point is completed; If traffic signs, traffic markings, safety facilities, technological facilities and other facilities are missing or have safety hazards under the best viewing angle, you need to go to Step 5 to confirm the details of traffic signs, traffic markings, safety facilities, technological facilities and other facilities; Step 5: Adjust according to the specifications Values ​​are used to confirm whether traffic signs, traffic markings, safety facilities, scientific and technological facilities, and other facilities are missing before and after the best inspection points: If adjusted If traffic signs, traffic markings, safety facilities, scientific and technological facilities and other facilities are still not found after the inspection, they will be deemed to be missing; If adjusted If you find traffic signs, traffic markings, safety facilities, scientific and technological facilities and other facilities after the inspection, you need to go to Step 6 to determine the specific hidden dangers; Step 6: Determine whether traffic signs, traffic markings, safety facilities, scientific and technological facilities and other facilities are complete: If they are complete, it is determined that the locations of traffic signs, traffic markings, safety facilities, scientific and technological facilities and other facilities do not meet the standards; Increase the recognition setting position is too far from the reference point, Reduce the distance between the identified setting position and the reference point; If it is incomplete, combine it with Step 5 to comprehensively give the hidden danger identification results; Step 7: Output the hidden danger identification results in the form of picture annotations and text summaries to complete the intelligent investigation of safety hazards; Step 8: Form conclusions on the inspection of road traffic safety hazards in tunnel sections, including a report on the inspection process of road traffic safety hazards in tunnel sections, a report on road traffic safety hazards in tunnel sections, and an overall summary report based on the inspection of hazards, and summarize the inspection results of traffic signs, traffic markings, safety facilities, scientific and technological facilities and other facilities in tunnel sections.

[0017] In the above technical scheme, the technical effects and advantages provided by the present invention are as follows: First, by integrating a variety of front-end intelligent devices such as high-definition cameras, GPS measuring instruments, inertial navigation instruments, etc., it is possible to efficiently collect multi-dimensional data of roads, including key traffic information such as traffic signs, traffic markings, and safety facilities, to ensure the standardization and accuracy of the data. Secondly, the present invention uses AI algorithms to automatically identify kilometer piles and hundred-meter piles on the collected images, and combines manual auxiliary confirmation to achieve accurate calibration of the starting and ending points and the up and down directions of the tunnel section. In addition, through the construction of standardized data and automated cleaning technology, duplicate and redundant data are effectively reduced, and the accuracy of the investigation is improved. The application of laser radar technology not only improves the ability to detect obstacles, but also ensures real-time performance and response speed, and can intelligently identify and quickly handle potential safety hazards. On the whole, the present invention realizes comprehensive and systematic identification of traffic safety hazards in tunnel sections through standardized data processing, intelligent identification technology, and a full range of hidden danger investigation processes, effectively improving the safety management level of road traffic. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present application 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 recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0019] Figure 1 It is a schematic diagram of the technical route of the tunnel section road traffic safety hazard inspection process of the present invention.

[0020] Figure 2 It is a schematic diagram of the flow chart of the tunnel section data discrimination algorithm of the present invention.

[0021] Figure 3 It is a schematic diagram of the technical route of the tunnel section safety hazard inspection method of the present invention. DETAILED DESCRIPTION

[0022] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of the present disclosure will be more comprehensive and complete, and the concept of the example embodiments will be fully conveyed to those skilled in the art.

[0023] The present invention provides Figures 1 to 3 A method for troubleshooting road traffic safety hazards based on a tunnel section scenario is shown, comprising the following steps: S1. Construct a knowledge system for checking road traffic safety hazards in tunnel sections; S2. Establish a standardized data collection process; S3, cleaning the road section data based on the map; S4, perform systematic stake number and inspection section calibration; S5. Establish intelligent tunnel section identification; S6. Standardized inspection method and conclusion of road traffic safety hazards in tunnel sections; In S1, the specific steps are as follows: In S1, the specific steps are as follows: 1.1. Collect basic parameters and structural information of the tunnel section.

[0024] 1.2 Traffic signs and road markings: The system identifies and checks traffic signs and road markings inside and outside the tunnel, including: Traffic markings: Check the dividing lines of the up and down lanes, the edge lines of the lanes, and the vertical markings.

[0025] Traffic signs: Check upstream and downstream tunnel signs, headlight signs, height limit signs, width limit signs, speed limit signs, no overtaking signs, fire equipment signs, emergency telephone signs, evacuation signs, tunnel exit distance warning signs, emergency parking lane signs, and speed limit removal signs.

[0026] 1.3. Safety facilities: Check raised road signs, contour markers, tunnel contour strips, emergency parking strips, escape routes, emergency telephones, and fire-fighting facilities.

[0027] 1.4. Technological facilities: testing lane signal lights and monitoring equipment technological facilities.

[0028] 1.5. Other facilities: Check the lighting facilities and height limit frames inside and outside the tunnel.

[0029] The inspection of road traffic safety hazards in tunnel sections involves basic information, traffic signs, traffic markings, safety facilities, scientific and technological facilities and other facilities. The actual identification of road traffic safety hazards involves many laws and regulations. To achieve a comprehensive and detailed inspection of road traffic safety hazards in tunnel sections, it is necessary to rely on relevant road traffic safety standards and build a knowledge system for the inspection of road traffic safety hazards in tunnel sections. This knowledge system includes basic information about tunnel sections and the types of hazards that may exist in traffic signs, traffic markings, safety facilities, scientific and technological facilities and other facilities. It also includes hazard identification standards, location information of hazard inspection points, description of hazard content and treatment suggestions. The traffic markings checked in the tunnel section include the upstream and downstream opposite lane dividing lines, the same direction lane dividing lines, the lane edge lines, and the facade markings; the traffic signs checked include the upstream and downstream tunnel signs, headlight signs, height limit signs, width limit signs, speed limit signs, no overtaking signs, fire equipment signs, emergency telephone signs, evacuation signs, tunnel exit distance warning signs, emergency parking lane signs, emergency parking lane position warning signs, speed limit lifting signs and no overtaking lifting signs; safety facilities include raised road signs, contour signs, tunnel contour strips, emergency parking lanes, escape routes, emergency telephones, and fire-fighting facilities; scientific and technological facilities mainly focus on lane signal lights. In addition, attention should also be paid to the hidden dangers of other facilities such as lighting facilities and height limit frames.

[0030] The knowledge system for checking road traffic safety hazards in tunnel sections constructed in this paper is not limited to the identification of whether there are hidden dangers in tunnel sections. More importantly, it is to determine the location information of hidden danger points and sections in the section and analyze the hidden dangers, so as to ensure the integrity and standardization of the hidden danger inspection of traffic signs, traffic markings, safety facilities, scientific and technological facilities and other facilities in the tunnel sections. In S2, the specific steps are as follows: Equipment layout: Use multiple high-definition cameras installed on the modified vehicle to collect road image data at high frequency from the left front, front and right front perspectives; GPS information collection: Use the GPS measuring instrument to obtain real-time position data of the vehicle during driving; Vehicle status monitoring: Use the inertial navigation system to collect dynamic information of roll angle, pitch angle and yaw angle, record the spatial posture of the vehicle, combine the speed sensor to collect speed information, and use the altitude measuring instrument to obtain the altitude of the vehicle; Obstacle detection: Use laser radar to detect the front, left and right sides of the vehicle in real time, and record the existence and distance information of obstacles; Data storage and synchronization: All collected images, GPS, speed and altitude information are synchronized through time tags Manage, store in their respective equipment storage centers, and match and use in the troubleshooting process; based on intelligent front-end equipment such as on-board high-definition cameras, GPS measuring instruments, inertial navigation instruments, speed sensors, altitude measuring instruments, and laser radars, by deploying multiple high-definition cameras on modified vehicles, multi-view high-frequency road image data is collected from different directions such as the left front, the front, and the right front; use the GPS measuring instrument to collect GPS information during the vehicle's driving process; rely on the inertial navigation instrument to measure the roll angle, pitch angle, and yaw angle information; use the speed sensor to measure the speed information and the altitude measuring instrument to measure the altitude of the vehicle during driving; use the laser radar to detect the presence of tunnel sections directly above, on the left, and on the right of the vehicle, and the cross-sectional change scenes. After the above information is collected, it is stored in the corresponding equipment storage center respectively, and the time tag is used to complete the matching of different data features to construct standardized collection data, providing a data basis for subsequent tunnel section identification and safety hazard investigation; In S3, the specific steps are as follows: Data digitization processing: Using the video, GPS location, time and other information automatically collected by the vehicle-side equipment, the traffic signs and markings in the tunnel are digitized to generate corresponding image data; Data mapping: Upload digital road signs and road marking information to Baidu Maps and other map platforms based on their GPS data to create corresponding road atlases; Manual cleaning: For duplicate data generated during the collection process due to reasons such as waiting for red lights, manual cleaning is used to remove redundant or irrelevant image data to ensure that only the road section data that needs to be checked is retained; relying on the video, location, time and other related sign and marking data automatically collected by the vehicle-side equipment, the road sign and marking information is digitized. According to the GPS information of the picture, the data is mapped on Baidu Maps. When conducting daily inspections and special inspections, the collected road image data is likely to be repetitive, especially when the waiting time at the red light at the intersection is long. In order to ensure the uniqueness of the road atlas, the road section needs to be cleaned. This step uses manual cleaning. According to the inspection content, the sections that need special inspections are manually cleaned from the original data; In S4, the specific steps are as follows: 4.1. AI automatic identification: AI technology is used to identify kilometer posts and hundred-meter posts on the road, and each image is matched with its corresponding post number to generate an electronic post number; Direction calibration: The up and down directions of the tunnel section are calibrated by kilometer posts, and the images collected by the main camera and the right camera are used for auxiliary calibration; Manual confirmation: Based on AI recognition, the accuracy of the kilometer post calibration is further confirmed manually to ensure that the calibration result is consistent with the actual situation; Using the image of the right camera as a reference, the button calibration is performed manually on the image with kilometer post marks to generate the corresponding electronic post number; First, the image features are extracted through a convolutional neural network (CNN) to extract the feature vector :

[0031] in, is the input image data, is the extracted feature vector.

[0032] Identification of kilometer piles and hundred-meter piles: The extracted features Perform classification and use a trained deep learning model (such as ResNet or YOLO) to predict whether there are kilometer posts or hundred-meter posts, and output their specific locations:

[0033] in, Pile) is the recognition probability of kilometer pile or hundred-meter pile, and are the weights and biases of the neural network, is the activation function.

[0034] Manual confirmation algorithm: Use the image of the manually calibrated kilometer pile to confirm the calibration result: = +

[0035] in, is the final confirmed pile number. is the pile number identified by Al, This is the adjustment value after manual confirmation.

[0036] Stake Matching: Match the actual stake of the image:

[0037] in, is the generated electronic stake number, Representing images Sign with kilometer post The matching function of Up and down direction calibration: Through the identified kilometer pile or hundred-meter pile, use the image data of the main camera and the right camera to calculate the spatial position of the vehicle relative to the pile number and determine the up and down direction:

[0038] in, is the direction calibration result, and are the pitch and yaw angles of the vehicle attitude, Direction calibration function 4.2. Task creation: According to the requirements of the inspection task, determine the road section information that needs to be inspected; Determine the road section information: According to the collected image information, determine the name, region, jurisdiction, uplink and downlink directions, start and end point pile numbers and other basic data of each inspection section; Calibration content: Complete the calibration of key information such as the road section name, inspection mileage, data collection time, speed limit, etc.; Through the collected images , combined with GPS data Extract road segment information: ;in, For each image The corresponding road section information, For GPS data, It is a function to extract road segment information from images and GPS.

[0039] Matching and calibration of road section information: Using road section information in the prior knowledge base , the collected road section information Perform matching and calibration: ;in, For the finalized road section information, The collected road information and road segment information in the knowledge base Similarity matching function; Determination of calibration content: Check the mileage calibration: Calculate and check mileage using GPS data: , in, To check the mileage, dist For GPS data points and The distance between is the number of GPS data points; Data collection time calibration: Based on the timestamp of each image ,Calibrate the time interval of data collection: ; in, is the start collection time, The collection time ends.

[0040] In S5, the specific steps are as follows: when conducting the inspection of road traffic safety hazards in tunnel sections, before the inspection of tunnel sections, it is necessary to realize intelligent identification of tunnel sections that are not limited to constant road cross-sectional widths, determine the corresponding starting and ending position information of the tunnel sections, and combine the intelligent identification with front-end perception and manual assistance for dual identification; perform obstacle discrimination on the signal data returned by the front-end perception device, identify and calculate the basic parameters of the section, dynamically identify and perceive the tunnel section for the slight or differential changes in the road cross-sectional width of different tunnel sections, and use the tunnel section perception discrimination algorithm calculation and manual assistance to identify the tunnel section. The tunnel section discrimination method is as follows: the laser radar emits a laser beam at a certain frequency to detect whether there are walls, metals and other objects that are characteristics of the tunnel section directly above, on the left and on the right of the vehicle; compare the received signal reflected from the target with the transmitted signal to create a set of point clouds; measure the distance to the wall by calculating the time difference between emission and reflection; the tunnel section recognition algorithm is designed as follows: the laser radar emits a laser beam at a certain frequency to detect whether there are walls, metals and other objects that are characteristics of the tunnel section directly above, on the left and on the right of the vehicle. The received signal reflected from the target is compared to the transmitted signal to create a set of point clouds. The distance to the wall is measured by calculating the time difference between transmission and reflection.

[0041] According to the "Technical Standards for Highway Engineering" and "Highway Tunnel Design Specifications Volume 1 Civil Engineering", the reference standards for the inspection of safety hazards in tunnel sections are determined as follows: During the data collection process, in order to distinguish between the underpass section and the tunnel section, the maximum width of the cross section of the general integrated highway is 42m. Considering the existence of one-way and two-way traffic in the tunnel, the minimum measurable distance on the left side is defined as 0.5m, the same as the right side. (The width of the measuring vehicle is 1.8m, the width of the radar equipment is 0.8m, and the minimum distance from the laser radar emission point to the vehicle body boundary on both sides is 0.5m.) The construction height limit of the tunnel cross section is 5.0m for expressways, first-class highways, and second-class highways, and 4.5m for third-class and fourth-class highways. Since the minimum construction height limit is 4.5m, the minimum detection distance directly above is set to 2.5m to prevent interference from other objects above. (The detector beam emission point is set to 2m from the ground) According to 1.0.4 of Highway Tunnel Design Code, Volume 1, Civil Engineering (JTG 3370.1-2018), tunnels are divided into four categories according to their length: extra-long tunnels, long tunnels, medium tunnels and short tunnels.

[0042] Classification Extra long tunnel Long Tunnel Middle Tunnel Short tunnel Length (m) L＞3000 3000≥L＞1000 1000≥L＞500 L≤500 Step 1: Automatic calculation time The reflection distance of the radar receiver directly above the point is , where the reflection distance calculation formula is , c represents the propagation speed of laser in air, T represents the time interval from light beam emission to reception; When there is no reflection, T is 0 Step2: Judgment Is between 2.5 and 15: like <2.5 or >15, the point does not belong to the tunnel section; If 2.5≤ ≤15, go to Step 3; Step 3: Calculation time Point right radar receiver reflection distance ; Step4: Judgment Is it greater than 0.5? like ≤0.5, the point does not belong to the tunnel section; like >0.5, then go to Step 5; Step 5: Calculation time Point left radar receiver reflection distance ; Step 6: Judgement Is it greater than 0.5? like ≤0.5, the point does not belong to the tunnel section; like >0.5, the point may be in a tunnel section, record the starting point Point is the starting point of the tunnel section , go to Step 7; Step 7: Order ,calculate time The reflection distance of the radar receiver directly above the point is ,in is the data collection time interval; Step 8: Judgement Is between 2.5 and 15: like <2.5 or >15, the point does not belong to the tunnel section; If 2.5≤ ≤15, go to Step 9; Step 9: Calculation time Point right radar receiver reflection distance ; Step10: Judgment Is it greater than 0.5? like ≤0.5, the point does not belong to the tunnel section; like >0.5, then go to Step 11; Step 11: Calculation time Point left radar receiver reflection distance ; Step12: Judgment Is it greater than 0.5? like ≤0.5, the point does not belong to the tunnel section; like >0.5, the point may be in a tunnel section. Point and tunnel section start Distance , go to Step 13; Step13: Judgment Is it greater than 42m? like <42m, then the road section may be under a bridge; like ≥42m, the road section is in a tunnel section; Step 14: Construct a set of points in the tunnel section in chronological order ,definition and are the starting point and end point of the tunnel section respectively, and go to Step 15; Step15: Judgment Is it greater than 3000m? like <3000m, the tunnel does not belong to the extra-long tunnel section, and the process goes to Step 16; like ≥3000m, the tunnel is a special long tunnel section; Step16: Judgment Is it greater than 1000m? like <1000m, the tunnel does not belong to the long tunnel section, and go to Step 17; like ≥1000m, the tunnel is a long tunnel section; Step17: Judgment Is it greater than 500m? like <500m, the tunnel is a short tunnel section; like ≥500m, the tunnel belongs to the middle tunnel section; The tunnel section data discrimination algorithm is used to identify tunnel sections with not limited to constant cross-sectional width and dynamically changing cross-sectional width, determine the starting and ending points, and identify and judge the tunnel length category.

[0043] In S6, the specific steps are as follows: The tunnel section road traffic safety hazard inspection method is based on the constructed safety hazard inspection knowledge system, and the tunnel section is inspected from different aspects of traffic signs, traffic markings, safety facilities, scientific and technological facilities and other facilities; firstly, the layout positions of the tunnel section traffic signs, traffic markings, safety facilities, etc. are inspected through the speed limit of the section in the basic information. After the basic information database of the section is constructed, the contents of the tunnel section traffic signs, traffic markings, safety facilities, scientific and technological facilities and other facilities can be clarified; the tunnel section inspection method relies on intelligently collected pictures to visually inspect the various hazard contents of the tunnel section; the inspection steps are as follows: Step 1: Determine the starting and ending points of the tunnel section ; Step 2: According to the standard provisions in the knowledge system for safety hazard inspection and the location information of safety hazards in the tunnel section, determine the best points for inspection of traffic signs, traffic markings, safety facilities, scientific and technological facilities and other facilities. ;in , The distance or range between the inspection points and the reference points for traffic signs, traffic markings, safety facilities, scientific and technological facilities and other facilities specified in relevant standards; Step 3: Determine the best point The picture below is the best angle for troubleshooting safety hazards; Step 4: Confirm whether traffic signs, traffic markings, safety facilities, scientific and technological facilities and other facilities are missing or whether there are safety hazards based on the image from the best viewing angle: If there are no safety hazards at the best viewing angle, the hidden danger inspection for this point is completed; If traffic signs, traffic markings, safety facilities, technological facilities and other facilities are missing or have safety hazards under the best viewing angle, you need to go to Step 5 to confirm the details of traffic signs, traffic markings, safety facilities, technological facilities and other facilities; Step 5: Adjust according to the specifications Values ​​are used to confirm whether traffic signs, traffic markings, safety facilities, scientific and technological facilities, and other facilities are missing before and after the best inspection points: If adjusted If traffic signs, traffic markings, safety facilities, scientific and technological facilities and other facilities are still not found after the inspection, they will be deemed to be missing; If adjusted If you find traffic signs, traffic markings, safety facilities, scientific and technological facilities and other facilities after the inspection, you need to go to Step 6 to determine the specific hidden dangers; Step 6: Determine whether traffic signs, traffic markings, safety facilities, scientific and technological facilities and other facilities are complete: If they are complete, it is determined that the locations of traffic signs, traffic markings, safety facilities, scientific and technological facilities and other facilities do not meet the standards; Increase the recognition setting position is too far from the reference point, Reduce the distance between the identified setting position and the reference point; If it is incomplete, combine it with Step 5 to comprehensively give the hidden danger identification results; Step 7: Output the hidden danger identification results in the form of picture annotations and text summaries to complete the intelligent investigation of safety hazards; Step 8: Form a conclusion on the road traffic safety hazard inspection of the tunnel section, including a road traffic safety hazard inspection process report on the tunnel section, a road traffic safety hazard report on the tunnel section, and an overall summary report based on the hazard inspection.

[0044] The intelligent inspection of safety hazards in tunnel sections combines intelligent algorithms with manual assistance to determine the best viewing angle for inspection of hidden dangers in tunnel sections, selects image-related data of target points for inspection of hidden dangers, first identifies hidden dangers using deep image recognition technology, and assists with manual judgment to ensure a comprehensive inspection of safety hazards in tunnel sections, and annotates hidden dangers discovered during the inspection in the form of picture tags, making it easier to attach the annotated hidden danger pictures to the inspection report and present the inspection results in a visual form.

[0045] In order to achieve hidden danger investigation conclusions for different users, this article defines a hierarchical investigation report: including tunnel section road traffic safety hidden danger investigation process report, tunnel section road traffic safety hidden danger report, and overall summary report based on hidden danger investigation. The tunnel section road traffic safety hidden danger investigation process report can summarize the investigation results of traffic signs, traffic markings, safety facilities, scientific and technological facilities and other facilities in the tunnel section, and its role is to ensure the integrity and correctness of the investigation process; the tunnel section road traffic safety hidden danger list and the overall report based on hidden danger investigation analyze the existing hidden danger content from part to whole, and its role is to view the hidden danger investigation conclusions in different permissions and scopes according to different user business needs. The above investigation reports defined in this article are based on the investigation report format requirements in the investigation work specifications, which is convenient for reporting the investigation conclusions to relevant departments during the road traffic safety hidden danger investigation work and for the subsequent safety hidden danger management.

[0046] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A method for checking potential road traffic safety hazards based on a tunnel section scenario, characterized in that: The following steps are involved: S1. Construct a knowledge system for checking road traffic safety hazards in tunnel sections; S2. Establish a standardized data collection process; S3, cleaning the road section data based on the map; S4, perform systematic stake number and inspection section calibration; S5. Establish intelligent tunnel section identification; S6. Standardized inspection method and conclusion of road traffic safety hazards in tunnel sections; In S1, the specific steps are as follows: 1.

1. Collect basic parameters and structural information of tunnel sections; 1.2 Traffic signs and road markings: The system identifies and checks traffic signs and road markings inside and outside the tunnel, including: Traffic markings: check the dividing lines of up and down lanes, lane edge lines, and vertical markings; Traffic signs: Check upstream and downstream tunnel signs, headlight signs, height limit signs, width limit signs, speed limit signs, no overtaking signs, fire equipment signs, emergency telephone signs, evacuation signs, tunnel exit distance warning signs, emergency parking strip signs, and speed limit removal signs; 1.

3. Safety facilities: Check raised road signs, contour signs, tunnel contour strips, emergency parking strips, escape routes, emergency telephones, and fire-fighting facilities; 1.

4. Scientific and technological facilities: testing lane signal lights, monitoring equipment and scientific and technological facilities; 1.

5. Other facilities: Check the lighting facilities and height limit frames inside and outside the tunnel; In S2, the specific steps are as follows: Equipment layout: Utilize multiple high-definition cameras installed on the modified vehicle to collect road image data at high frequency from the left front, front and right front perspectives; GPS information collection: Use the GPS measuring instrument to obtain real-time position data of the vehicle during driving; Vehicle status monitoring: Use the inertial navigation system to collect dynamic information of roll angle, pitch angle and yaw angle, record the spatial posture of the vehicle, combine the speed sensor to collect vehicle speed information, and use the altitude measuring instrument to obtain the altitude of the vehicle; Obstacle detection: Use laser radar to perform real-time detection of the front, left and right sides of the vehicle, and record the existence and distance information of obstacles; Data storage and synchronization: All collected images, GPS, vehicle speed and altitude information are synchronized through time tags, stored in their respective equipment storage centers, and matched and used during the troubleshooting process.

2. According to claim 1, a method for checking potential road traffic safety hazards based on a tunnel section scenario is characterized in that: In S3, the specific steps are as follows: Data digital processing: Utilize the video, GPS location, and time information automatically collected by the vehicle-side equipment to digitally process the traffic signs and markings in the tunnel and generate corresponding image data; Data mapping: Upload the digitized road signs and road markings to the Baidu map platform based on their GPS data to create a corresponding road atlas; Manual cleaning: For duplicate and redundant data generated during the collection process due to reasons such as waiting for red lights, manual cleaning is used to eliminate redundant or irrelevant image data to ensure that only the road section data that needs to be checked is retained.

3. According to claim 1, a method for checking potential road traffic safety hazards based on a tunnel section scenario is characterized in that: In S4, the specific steps are as follows: 4.

1. AI automatic identification: AI technology is used to identify kilometer posts and hundred-meter posts on the road, and each picture is matched with its corresponding post number to generate an electronic post number; Direction calibration: The up and down directions of the tunnel section are calibrated by kilometer posts, and the images collected by the main camera and the right camera are used for auxiliary calibration; Manual confirmation: Based on AI identification, the accuracy of the kilometer post calibration is further confirmed manually to ensure that the calibration result is consistent with the actual situation; Using the image of the right camera as a reference, manual button calibration is performed on the image with kilometer post marks to generate the corresponding electronic post number; 4.

2. Task creation: Determine the road section information that needs to be checked according to the requirements of the inspection task; Determine the road section information: According to the collected image information, determine the name, region, jurisdiction, up and down directions, and basic data of the starting and ending points of each road section to be checked; Calibration content: complete the calibration of key information such as road section name, inspection mileage, data collection time, and speed limit.

4. The method for checking road traffic safety hazards based on a tunnel section scenario according to claim 1 is characterized in that: In S5, the specific steps are as follows: when conducting the inspection of road traffic safety hazards in tunnel sections, it is necessary to realize intelligent identification of tunnel sections that are not limited to a constant road cross-sectional width before the inspection of the tunnel sections, determine the corresponding starting and ending position information of the tunnel sections, and combine intelligent identification with front-end perception and manual assistance for dual identification; Obstacle discrimination is performed on the signal data returned by the front-end sensing device, and the basic parameters of the road section are identified and calculated. The tunnel section is dynamically identified and perceived for the subtle or differential changes in the width of the road cross section of different tunnel sections. The tunnel section is identified by using the tunnel section perception discrimination algorithm calculation and manual assistance. The tunnel section discrimination method is as follows: The laser radar emits a laser beam at a certain frequency to detect whether there are walls, metals and other objects that are characteristic of the tunnel section directly above, on the left and on the right of the vehicle; the received signal reflected from the target is compared with the transmitted signal to create a set of point clouds; the distance to the wall is measured by calculating the time difference between emission and reflection; the tunnel section recognition algorithm is designed as follows: Step 1: Automatic calculation time The reflection distance of the radar receiver directly above the point is , where the reflection distance calculation formula is , c represents the propagation speed of laser in air, T represents the time interval from light beam emission to reception; When there is no reflection, T is 0 Step2: Judgment Is between 2.5 and 15: like <2.5 or >15, the point does not belong to the tunnel section; If 2.5≤ ≤15, go to Step 3; Step 3: Calculation time Point right radar receiver reflection distance ; Step4: Judgment Is it greater than 0.5? like ≤0.5, the point does not belong to the tunnel section; like >0.5, then go to Step 5; Step 5: Calculation time Point left radar receiver reflection distance ; Step 6: Judgement Is it greater than 0.5? like ≤0.5, the point does not belong to the tunnel section; like >0.5, the point may be in a tunnel section, record the starting point Point is the starting point of the tunnel section , go to Step 7; Step 7: Order ,calculate time The reflection distance of the radar receiver directly above the point is ,in is the data collection time interval; Step 8: Judgement Is between 2.5 and 15: like <2.5 or >15, the point does not belong to the tunnel section; If 2.5≤ ≤15, go to Step 9; Step 9: Calculation time Point right radar receiver reflection distance ; Step10: Judgment Is it greater than 0.5? like ≤0.5, the point does not belong to the tunnel section; like >0.5, then go to Step 11; Step 11: Calculation time Point left radar receiver reflection distance ; Step12: Judgment Is it greater than 0.5? like ≤0.5, the point does not belong to the tunnel section; like >0.5, the point may be in a tunnel section. Point and tunnel section start Distance , go to Step 13; Step13: Judgment Is it greater than 42m? like <42m, then the road section may be under a bridge; like ≥42m, the road section is in a tunnel section; Step 14: Construct a set of points in the tunnel section in chronological order ,definition and The starting point and end point of the tunnel section are shown respectively, and then go to Step 15; Step15: Judgment Is it greater than 3000m? like <3000m, the tunnel does not belong to the extra-long tunnel section, and the process goes to Step 16; like ≥3000m, the tunnel is a special long tunnel section; Step16: Judgment Is it greater than 1000m? like <1000m, the tunnel does not belong to the long tunnel section, and go to Step 17; like ≥1000m, the tunnel is a long tunnel section; Step17: Judgment Is it greater than 500m? like <500m, the tunnel is a short tunnel section; like ≥500m, the tunnel belongs to the middle tunnel section.

5. The method for checking road traffic safety hazards based on a tunnel section scenario according to claim 1 is characterized in that: In S6, the specific steps are as follows: The tunnel section road traffic safety hazard inspection method is based on the constructed safety hazard inspection knowledge system, and the tunnel section is inspected from different aspects of traffic signs, traffic markings, safety facilities, scientific and technological facilities and other facilities; firstly, the traffic signs, traffic markings, and safety facilities of the tunnel section are inspected through the speed limit of the section in the basic information. After the basic information database of the section is constructed, the traffic signs, traffic markings, safety facilities, scientific and technological facilities and other facilities of the tunnel section can be clearly defined; the tunnel section inspection method relies on intelligently collected pictures to visually inspect the various hazard contents of the tunnel section; the inspection steps are as follows: Step 1: Determine the starting and ending points of the tunnel section ; Step 2: According to the standard provisions in the knowledge system for safety hazard inspection and the location information of safety hazards in the tunnel section, determine the best points for inspection of traffic signs, traffic markings, safety facilities, scientific and technological facilities and other facilities. ; ,in, The distance or range between the inspection points and the reference points for traffic signs, traffic markings, safety facilities, scientific and technological facilities and other facilities specified in relevant standards; Step 3: Determine the best point The picture below is the best angle for troubleshooting safety hazards; Step 4: Confirm whether traffic signs, traffic markings, safety facilities, scientific and technological facilities and other facilities are missing or whether there are safety hazards based on the image from the best viewing angle: If there are no safety hazards at the best viewing angle, the hidden danger inspection for this point is completed; If traffic signs, traffic markings, safety facilities, technological facilities and other facilities are missing or have safety hazards under the best viewing angle, you need to go to Step 5 to confirm the details of traffic signs, traffic markings, safety facilities, technological facilities and other facilities; Step 5: Adjust according to the specifications Values ​​are used to confirm whether traffic signs, traffic markings, safety facilities, scientific and technological facilities, and other facilities are missing before and after the best inspection points: If adjusted If traffic signs, traffic markings, safety facilities, scientific and technological facilities and other facilities are still not found after the inspection, they will be deemed to be missing; If adjusted If you find traffic signs, traffic markings, safety facilities, scientific and technological facilities and other facilities after the inspection, you need to go to Step 6 to determine the specific hidden dangers; Step 6: Determine whether traffic signs, traffic markings, safety facilities, scientific and technological facilities and other facilities are complete: If they are complete, it is determined that the locations of traffic signs, traffic markings, safety facilities, scientific and technological facilities and other facilities do not meet the standards; Increase the recognition setting position is too far from the reference point, Reduce the distance between the identified setting position and the reference point; If it is incomplete, combine it with Step 5 to comprehensively give the hidden danger identification results; Step 7: Output the hidden danger identification results in the form of picture annotations and text summaries to complete the intelligent investigation of safety hazards; Step 8: Form conclusions on the inspection of road traffic safety hazards in tunnel sections, including a report on the inspection process of road traffic safety hazards in tunnel sections, a report on road traffic safety hazards in tunnel sections, and an overall summary report based on the inspection of hazards, and summarize the inspection results of traffic signs, traffic markings, safety facilities, scientific and technological facilities and other facilities in tunnel sections.

Images