Method and system for evaluating the impact of tunnel structure damage repair on driving safety

Abstract

The application discloses a kind of tunnel structure damage repair to the evaluation method and system of influence on driving safety, comprising: constructing the tunnel repair scene after tunnel structure damage repair;Collect the visual characteristic information when driving personnel passes through the tunnel repair scene;According to the visual characteristic information, the regional visual characteristic value of driving personnel is calculated;The influence on driving safety is evaluated using the regional visual characteristic value, the greater the regional visual characteristic value, the smaller the influence on driving safety after tunnel structure damage repair;The smaller the regional visual characteristic value, the greater the influence on driving safety after tunnel structure damage repair.The application can quantify the influence of reinforcement and repair disposal result under different damage modes of tunnel on driving safety of driver, and guarantee the normal and safe operation of tunnel traffic.

Description

Technical Field

[0001] This invention relates to the field of tunnel traffic safety, specifically to a method and system for assessing the impact of tunnel structural damage repair on traffic safety. Background Technology

[0002] With the rapid development of highway tunnels in my country, tunnel engineering has transitioned from construction to operation and maintenance. Problems such as surrounding rock cracking, lining spalling, structural cracks, and groundwater seepage have gradually become prominent. Moreover, emergencies occur frequently in tunnels, and the operational problems of "nine out of ten tunnels leaking and ten out of ten tunnels cracking" are not optimistic.

[0003] Tunnel structural damage modes are diverse, and there are numerous corresponding repair and reinforcement methods. The combined effect of different repair and reinforcement methods and tunnel structural damage modes alters the appearance of the tunnel structure. When drivers and passengers pass through the repaired section of the tunnel, their travel characteristics will be changed by the changes in the appearance of the tunnel structure, which will affect driving safety inside the tunnel.

[0004] Currently, there is a lack of testing technology to assess the impact of tunnel structural damage repair on traffic safety, making it difficult to effectively evaluate the impact of reinforcement and repair treatments on traffic safety under different tunnel structural damage modes. Therefore, a method and system for assessing the impact of tunnel structural damage repair on traffic safety is needed to achieve an effective evaluation. Summary of the Invention

[0005] In view of this, the purpose of this invention is to overcome the deficiencies in the prior art and provide a method and system for assessing the impact of tunnel structural damage repair on driving safety. This method and system can quantify the impact of reinforcement and repair treatments under different tunnel damage modes on driver safety and ensure the normal and safe operation of tunnel traffic.

[0006] The method for assessing the impact of tunnel structure damage repair on traffic safety according to the present invention includes:

[0007] Construct a tunnel repair scenario after structural damage has been repaired;

[0008] Collect visual characteristic information of the driver when passing through the tunnel repair scene;

[0009] Based on the visual characteristic information, calculate the driver's regional visual characteristic value;

[0010] The visual characteristic values ​​of the area are used to assess the impact on driving safety. The larger the visual characteristic value of the area, the smaller the impact on driving safety after the tunnel structure damage is repaired; the smaller the visual characteristic value of the area, the larger the impact on driving safety after the tunnel structure damage is repaired.

[0011] Furthermore, the visual characteristic information includes a visual characteristic heatmap.

[0012] Furthermore, based on the aforementioned visual characteristic information, the driver's regional visual characteristic value is calculated, specifically including:

[0013] The visual area of ​​a driver passing through a normal tunnel is divided to obtain a visual area map; wherein the visual area map includes M visual areas;

[0014] The visual region map and the visual characteristic heatmap are processed into grayscale to obtain the visual region grayscale map G. RQ And visual characteristic thermal grayscale image G RR ;

[0015] For M visual regions, calculate the grayscale image G corresponding to each visual region. RQ With grayscale image G RR The average similarity is used to obtain M average similarities;

[0016] Calculate the driver's regional visual characteristic value K based on the M average similarities:

[0017]

[0018] Where, k i SM represents the difference coefficient for the i-th visual region. i Let be the average similarity corresponding to the i-th visual region.

[0019] Furthermore, the visual area of ​​a driver passing through a normal tunnel scene is divided to obtain a visual area map, which specifically includes:

[0020] The area where the driver's visual focus is located is designated as the focal area;

[0021] Define the visual area that connects to the focal area as the connecting area;

[0022] Visual areas not connected to the focal area are divided into the first adjacent area and the second adjacent area according to their distance from the focal area from the closest to the farthest.

[0023] Furthermore, the visual region map and the visual characteristic heatmap are processed into grayscale according to the following formulas:

[0024] Gray=(R·30+G·59+B·11+50) / 100;

[0025] Where Gray represents the grayscale value of the image; R represents the red pixel value of the pixel; G represents the green pixel value of the pixel; and B represents the blue pixel value of the pixel.

[0026] Furthermore, the grayscale image G corresponding to each visual region is calculated according to the following formula. RQ With grayscale image G RR Average similarity:

[0027] SM(x,y)=I(x,y)·C(x,y)·S(x,y);

[0028] Where SM(x,y) represents the grayscale image G corresponding to pixel (x,y) in the target visual region. RQ With grayscale image G RR Average similarity;

[0029] I(x,y) represents the brightness contrast value; μ x Representing image G RQ The mean value of a pixel; μ y For image G RR The average value of the pixels; C1 is a constant;

[0030] C(x,y) represents the contrast ratio; δ x Representing image G RQ Standard deviation of pixels; δ y For image G RR The standard deviation of pixels; C2 is a constant;

[0031] S(x,y) represents the structural contrast value; δ xy Representing image G RQ With image G RR The covariance; C3 is a constant.

[0032] Furthermore, the difference coefficient k of the i-th visual region is determined according to the following formula. i :

[0033]

[0034] Where x represents the distance between the driver's i-th visual region and the visual focus region.

[0035] An assessment system for the impact of tunnel structure damage repair on traffic safety includes a scenario construction unit, an information acquisition unit, and a safety impact assessment unit.

[0036] The scene construction unit is used to construct a tunnel repair scene after the tunnel structure damage has been repaired;

[0037] The information acquisition unit is used to collect visual characteristic information of the driver when passing through the tunnel repair scene;

[0038] The safety impact assessment unit is used to calculate the driver's regional visual characteristic value based on the visual characteristic information; and to assess the impact on driving safety using the regional visual characteristic value. The larger the regional visual characteristic value, the smaller the impact on driving safety after the tunnel structure damage is repaired; the smaller the regional visual characteristic value, the larger the impact on driving safety after the tunnel structure damage is repaired.

[0039] Furthermore, based on the aforementioned visual characteristic information, the driver's regional visual characteristic value is calculated, specifically including:

[0040] The visual area of ​​a driver passing through a normal tunnel is divided to obtain a visual area map; wherein the visual area map includes M visual areas;

[0041] The visual region map and the visual characteristic heatmap are processed into grayscale to obtain the visual region grayscale map G. RQ And visual characteristic thermal grayscale image G RR ;

[0042] For M visual regions, calculate the grayscale image G corresponding to each visual region. RQ With grayscale image G RR The average similarity is used to obtain M average similarities;

[0043] Calculate the driver's regional visual characteristic value K based on the M average similarities:

[0044]

[0045] Where, k i SM represents the difference coefficient for the i-th visual region. i Let be the average similarity corresponding to the i-th visual region.

[0046] Furthermore, the grayscale image G corresponding to each visual region is calculated according to the following formula. RQ With grayscale image G RR Average similarity:

[0047] SM(x,y)=I(x,y)·C(x,y)·S(x,y);

[0048] Where SM(x,y) represents the grayscale image G corresponding to pixel (x,y) in the target visual region. RQ With grayscale image G RR Average similarity;

[0049] I(x,y) represents the brightness contrast value; μ x Representing image G RQ The mean value of a pixel; μ y For image GRR The average value of the pixels; C1 is a constant;

[0050] C(x,y) represents the contrast ratio; δ x Representing image G RQ Standard deviation of pixels; δ y For image G RR The standard deviation of pixels; C2 is a constant;

[0051] S(x,y) represents the structural contrast value; δ xy Representing image G RQ With image G RR The covariance; C3 is a constant.

[0052] The beneficial effects of this invention are as follows: This invention discloses a method and system for assessing the impact of tunnel structure damage repair on traffic safety. By constructing a tunnel structure damage model and applying different reinforcement and repair methods to the model, a virtual scene of tunnel structure damage repair is obtained. By analyzing simulated passage by different drivers in the virtual tunnel structure damage repair scene, the visual characteristics of different drivers in the virtual scene are obtained. A convolutional neural network is used to classify the drivers' visual characteristics into levels and calculate the regional visual characteristic values ​​of the drivers. The regional visual characteristic values ​​are then used to quantitatively assess the impact on traffic safety, providing technical support for effectively assessing the impact of repaired tunnels on traffic safety and ensuring the normal and safe operation of tunnel traffic. Attached Figure Description

[0053] The present invention will be further described below with reference to the accompanying drawings and embodiments:

[0054] Figure 1 This is a schematic diagram of the driving safety impact assessment method of the present invention;

[0055] Figure 2 This is a schematic diagram of the virtual tunnel damage state scenario of the present invention;

[0056] Figure 3 This is a schematic diagram of the virtual tunnel after accident recovery scenario according to the present invention;

[0057] Figure 4 This is a schematic diagram of the virtual tunnel arch repair scenario of the present invention;

[0058] Figure 5 This is a schematic diagram of a virtual tunnel steel plate strip repair scenario according to the present invention;

[0059] Figure 6 This is a schematic diagram of a virtual tunnel corrugated steel repair scenario according to the present invention;

[0060] Figure 7 This is a schematic diagram of a virtual tunnel grouting repair scenario according to the present invention;

[0061] Figure 8 This is a schematic diagram of the visual region segmentation results of the present invention;

[0062] Figure 9 This is a schematic diagram illustrating the driver's visual area according to the present invention;

[0063] Figure 10 This is a driver's visual area diagram according to the present invention;

[0064] Figure 11 This is a heat map of the driver's visual characteristics according to the present invention. Detailed Implementation

[0065] The present invention will be further described below with reference to the accompanying drawings, as shown in the figures:

[0066] The method for assessing the impact of tunnel structure damage repair on traffic safety according to the present invention includes the following steps:

[0067] Construct a tunnel repair scenario after structural damage has been repaired;

[0068] Collect visual characteristic information of drivers when they pass through the tunnel repair scene; the visual characteristic information includes a visual characteristic heat map.

[0069] Based on the visual characteristic information, calculate the driver's regional visual characteristic value;

[0070] The visual characteristic values ​​of the area are used to assess the impact on driving safety. The larger the visual characteristic value of the area, the smaller the impact on driving safety after the tunnel structure damage is repaired; the smaller the visual characteristic value of the area, the larger the impact on driving safety after the tunnel structure damage is repaired.

[0071] In this embodiment, 3D modeling software such as 3DMax is used to construct the main tunnel structure. Different damage conditions are added to render the main tunnel structure, obtaining virtual tunnel scenes under different working conditions. The virtual tunnel scenes simulate real tunnel conditions; the structures and electromechanical facilities within the tunnel are also reproduced in the real tunnel scene. The results of constructing different virtual tunnel scenes are shown below. Figures 2-7 As shown.

[0072] The constructed virtual tunnel scenario is integrated into a vehicle-in-the-loop simulation test system. Since the driver's reaction to the simulated tunnel damage and reinforcement / repair scenarios decreases over time, this invention adjusts the order of the test scenarios and then conducts a virtual reality test involving the driver and vehicle. The specific steps are as follows:

[0073] 1) Drivers familiarize themselves with the driving environment in the vehicle-in-the-loop simulation test system;

[0074] 2) Drivers travel through a virtual scene of a normally operating tunnel at speed limits of 80km / h and 100km / h respectively, and visual characteristic data of drivers are collected using an eye tracker.

[0075] 3) Considering the driver's acceptance of new things, conduct virtual scenario tests with minimal changes to the appearance of the tunnel structure. Drivers travel in virtual scenarios such as tunnel cracks, tunnel lining collapse, groundwater leakage and concrete delamination under speed limits of 80km / h and 100km / h respectively. Eye trackers are used to collect the driver's visual characteristic data.

[0076] 4) Tunnel reinforcement and repair aims to restore the tunnel to its original state as much as possible while ensuring safety. Therefore, a virtual scenario test of tunnel reinforcement and repair was conducted. Drivers traveled in virtual scenarios such as tunnel arch repair, tunnel corrugated steel repair, tunnel steel plate strip repair and tunnel grouting repair at speed limits of 80km / h and 100km / h respectively. Eye trackers were used to collect visual characteristic data of the drivers.

[0077] 5) Conduct virtual scenario tests after rapid reopening of tunnels after accidents. Drivers travel in virtual scenarios such as tunnel fire, tunnel explosion and tunnel collapse at speed limits of 80km / h and 100km / h, respectively. Eye trackers are used to collect visual characteristic data of drivers.

[0078] Among them, the in-loop simulation test system is used to combine real vehicles or vehicle subsystems with a simulated environment to conduct various tests, verifications and simulations. It is an existing system and will not be described in detail here.

[0079] In this embodiment, calculating the driver's regional visual characteristic value based on the visual characteristic information specifically includes:

[0080] The visual area of ​​a driver passing through a normal tunnel is divided to obtain a visual area map; wherein the visual area map includes M visual areas;

[0081] The visual region map and the visual characteristic heatmap are processed into grayscale to obtain the visual region grayscale map G. RQ And visual characteristic thermal grayscale image G RR ;

[0082] For M visual regions, calculate the grayscale image G corresponding to each visual region. RQ With grayscale image G RR The average similarity is used to obtain M average similarities;

[0083] Calculate the driver's regional visual characteristic value K based on the M average similarities:

[0084]

[0085] Where, k i SM represents the difference coefficient for the i-th visual region. i Let be the average similarity corresponding to the i-th visual region.

[0086] By conducting simulation experiments in a virtual tunnel scenario, visual characteristic data of drivers were collected. Combined with the characteristics of drivers' visual gaze area, duration of gaze in a certain area, number of gazes, and saccade amplitude, the visual characteristics of drivers during passage in a virtual tunnel scenario were analyzed.

[0087] The visual region of a driver passing through a normal tunnel is divided into segments to obtain a visual region map, which specifically includes:

[0088] The area where the driver's visual focus is located is designated as the focal area;

[0089] Define the visual area that connects to the focal area as the connecting area;

[0090] Visual areas not connected to the focal area are divided into the first adjacent area and the second adjacent area according to their distance from the focal area from the closest to the farthest.

[0091] In other words, taking the driver's normal visual focus as the central area, the overall visual range is divided into 12 parts, as shown in the following figure. Figure 8 As shown.

[0092] Specifically, such as Figure 9 As shown, when a driver is passing through a normal tunnel, their visual focus is a distance directly in front of them, which is the R6 area. When a driver is passing through a virtual tunnel scene, their gaze area is the R6 area, and if their visual focus is within the normal driving visual area, then the driver is considered to be passing safely.

[0093] When a driver is traveling through a virtual tunnel scene, if the area of ​​focus is R2, R5, R7, or R10 and the driver's visual focus deviates slightly from the normal driving visual area, then the driver's passage is considered to have a low risk.

[0094] When a driver is traveling through a virtual tunnel scene, if the area of ​​focus is R1, R3, R9, or R11 and the driver's visual focus deviates significantly from the normal driving visual area, then the driver's passage is considered to pose a significant risk.

[0095] When a driver is traveling through a virtual tunnel scene, if their gaze is focused on areas R4, R8, and R12, and their visual focus deviates significantly from the normal driving visual area, then the driver's passage is considered relatively dangerous.

[0096] Existing methods for calculating driver visual characteristics only extract partial data from the driver's visual region, resulting in data waste. This invention utilizes an eye tracker to acquire eye-tracking image data and dynamic perspective image data of the driver navigating different virtual tunnel scenarios. Through image partitioning difference analysis, it analyzes the driver's visual characteristics within different visual regions. By comprehensively considering the differences in driver visual characteristics across different visual regions, it classifies the impact on driving safety after structural damage repair in the operating tunnel.

[0097] When drivers pass through tunnels, the differences in their different visual regions have varying impacts on driving safety. While image difference analysis can comprehensively reflect a driver's visual characteristics, the degree to which these differences affect driving safety is difficult to determine. Based on the division of regions, the distance relationship between the driver's gaze area and the visual focal point area is analyzed. Let the distance between the visual focal point area and the driver be 0, the distance between the visual focal point area and an adjacent area be 1, the distance between adjacent areas be 2, and so on.

[0098] The closer the driver's gaze area is to the visual focal area, the safer their driving behavior; conversely, the farther apart they are, the more dangerous it is. Therefore, this invention constructs a visual region difference coefficient, and determines the difference coefficient k of the i-th visual region according to the following formula. i :

[0099]

[0100] Where x represents the distance between the driver's i-th visual region and the visual focus region.

[0101] In this embodiment, based on the driver's visual characteristic data and visual region data collected by the eye tracker, the visual region is divided into 3x4 matrix blocks. Image segmentation is used to segment different matrix blocks, and each matrix block is labeled with a name. The result is as follows: Figure 10 and Figure 11 As shown.

[0102] Since both the driver's visual region map and the visual characteristic heatmap are color RGB images with complex image elements, this invention converts the acquired RGB images into grayscale images. This reduces the computational load and effectively extracts features such as contours and textures, facilitating subsequent image comparison. The expression for converting an RGB image to a grayscale image is:

[0103] Gray=(R·30+G·59+B·11+50) / 100;

[0104] In the formula, Gray represents the grayscale value of the image; R represents the red pixel value of the pixel; G represents the green pixel value of the pixel; and B represents the blue pixel value of the pixel.

[0105] The driver's visual region map and visual characteristic heat map are converted into grayscale images G respectively. RQ ={RQ1,RQ2,...,RQ 12} and G RR ={RR1,RR2,...,RR 12}, calculate G respectively RQ With G RR The similarity of grayscale images in different regions comprehensively reflects the changes in the driver's visual characteristics.

[0106] By comparing image G RQ With G RR The brightness, contrast, and structure were comprehensively compared, and the visual characteristics of different areas of the driver's vision were analyzed. The expressions for brightness contrast, contrast, and structure are as follows:

[0107]

[0108] In the formula, I(x,y) represents the brightness contrast value; μ x Representing image G RQ The mean value of a pixel; μ y For image G RR The average value of the pixels; C1 is a constant, with a default value of 0.2.

[0109]

[0110] In the formula, C(x,y) represents the contrast ratio; δ x Representing image G RQ Standard deviation of pixels; δ y For image G RR The standard deviation of pixels; C2 is a constant, with a default value of 0.3.

[0111]

[0112] In the formula, S(x,y) represents the structural contrast value; δ x Representing image G RQ Standard deviation of pixels; δ y For image G RR Standard deviation of pixels; δ xy Representing image G RQ With image G RRThe covariance; C3 is a constant, defaulting to 0.2.

[0113] Then the grayscale image G corresponding to each visual region can be obtained. RQ With grayscale image G RR Average similarity:

[0114] SM(x,y)=I(x,y)·C(x,y)·S(x,y);

[0115] In the formula, SM(x,y) represents the grayscale image G corresponding to pixel (x,y) in the target visual region. RQ With grayscale image G RR The average similarity; where pixel (x,y) can be the center pixel of the target visual region or the pixel corresponding to the average pixel value of the target visual region; the target visual region can be the i-th visual region, then SM(x,y) is the average similarity SM corresponding to the i-th visual region. i .

[0116] The average similarity of each region in the driver's visual region map and visual characteristic heatmap was calculated separately, resulting in the average similarity of 12 regions. Since the differences between different visual regions have varying degrees of impact on driving safety, the driver's regional visual characteristic value was comprehensively calculated based on the visual region difference coefficients of different regions. The expression for this value is as follows:

[0117]

[0118] In the formula, i represents the i-th visual region; k i This represents the difference coefficient of the i-th visual region.

[0119] The larger K is, the more concentrated the driver's visual characteristics are in front of the vehicle, and the less impact the repair of the tunnel structure damage will have on driving safety.

[0120] The smaller the value of K, the more dispersed the driver's visual characteristics are, the more visual information they receive, and the greater the impact on driving safety after the structural damage of the operating tunnel is repaired.

[0121] This invention also relates to an assessment system for the impact of tunnel structure damage repair on traffic safety. The system corresponds to the above-mentioned assessment method for the impact of tunnel structure damage repair on traffic safety and can be understood as a system for implementing the above method. The system includes a scenario construction unit, an information acquisition unit, and a safety impact assessment unit.

[0122] The scene construction unit is used to construct a tunnel repair scene after the tunnel structure damage has been repaired;

[0123] The information acquisition unit is used to collect visual characteristic information of the driver when passing through the tunnel repair scene;

[0124] The safety impact assessment unit is used to calculate the driver's regional visual characteristic value based on the visual characteristic information; and to assess the impact on driving safety using the regional visual characteristic value. The larger the regional visual characteristic value, the smaller the impact on driving safety after the tunnel structure damage is repaired; the smaller the regional visual characteristic value, the larger the impact on driving safety after the tunnel structure damage is repaired.

[0125] This invention provides an objective and scientific method and system for assessing the impact of tunnel structural damage repair on traffic safety. It enables effective assessment of the impact of tunnel structural damage repair on traffic safety, and can quantify the impact of reinforcement and repair treatments under different tunnel damage modes on driver safety, ensuring the safe and reliable operation of tunnel traffic and reducing the occurrence of tunnel traffic accidents.

[0126] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for assessing the impact of tunnel structural damage repair on traffic safety, characterized in that: include: Construct a tunnel repair scenario after structural damage has been repaired; Collect visual characteristic information of the driver when passing through the tunnel repair scene; The visual characteristic information includes a visual characteristic heatmap; Based on the aforementioned visual characteristic information, the driver's regional visual characteristic value is calculated, specifically including: The visual region of a driver navigating a normal tunnel is divided to obtain a visual region map; wherein, the visual region map includes... One visual area; The visual region map and the visual characteristic heatmap are processed into grayscale to obtain the visual region grayscale map. and visual characteristic thermal grayscale image ; against For each visual region, calculate the corresponding grayscale image. With grayscale image The average similarity is obtained. Average similarity; according to Average similarity is used to calculate the driver's regional visual characteristic value. : ； in, For the first The difference coefficient of each visual region For the first Average similarity of visual regions; The visual characteristic values ​​of the area are used to assess the impact on driving safety. The larger the visual characteristic value of the area, the smaller the impact on driving safety after the tunnel structure damage is repaired; the smaller the visual characteristic value of the area, the greater the impact on driving safety after the tunnel structure damage is repaired.

2. The method for assessing the impact of tunnel structural damage repair on traffic safety according to claim 1, characterized in that: The visual region of a driver passing through a normal tunnel is divided into segments to obtain a visual region map, which specifically includes: The area where the driver's visual focus is located is designated as the focal area; Define the visual area that connects to the focal area as the connecting area; Visual areas not connected to the focal area are divided into the first adjacent area and the second adjacent area according to their distance from the focal area from the closest to the farthest.

3. The method for assessing the impact of tunnel structural damage repair on traffic safety according to claim 1, characterized in that: The visual region map and the visual characteristic heatmap are processed into grayscale using the following formulas: ； in, Represents the grayscale value of the image; This represents the red pixel value of a pixel; This represents the green pixel value of a pixel; This represents the blue pixel value of a pixel.

4. The method for assessing the impact of tunnel structural damage repair on traffic safety according to claim 1, characterized in that: The grayscale image corresponding to each visual region is calculated using the following formula. With grayscale image Average similarity: ； in, Represents pixels in the target visual region Corresponding grayscale image With grayscale image Average similarity; Indicates the brightness contrast value; ; Representing an image The average value of a pixel; For image The average value of a pixel; It is a constant; Indicates the contrast ratio; ; Representing an image Standard deviation of pixels; For image Standard deviation of pixels; It is a constant; Indicates the structural comparison value; ; Representing an image With images covariance; It is a constant.

5. The method for assessing the impact of tunnel structural damage repair on traffic safety according to claim 1, characterized in that: The number is determined according to the following formula. Difference coefficient of visual regions : ； in, Indicates the driver's number The distance between each visual area and the visual focal area.

6. A system for assessing the impact of tunnel structural damage repair on traffic safety, characterized in that: It includes a scenario construction unit, an information collection unit, and a security impact assessment unit; The scene construction unit is used to construct a tunnel repair scene after the tunnel structure damage has been repaired; The information acquisition unit is used to collect visual characteristic information of the driver when passing through the tunnel repair scene; the visual characteristic information includes a visual characteristic heat map. The safety impact assessment unit is used to calculate the driver's regional visual characteristic value based on the visual characteristic information; and to assess the impact on driving safety using the regional visual characteristic value. The larger the regional visual characteristic value, the smaller the impact on driving safety after the tunnel structure damage is repaired; the smaller the regional visual characteristic value, the greater the impact on driving safety after the tunnel structure damage is repaired. Based on the aforementioned visual characteristic information, the driver's regional visual characteristic value is calculated, specifically including: The visual region of a driver navigating a normal tunnel is divided to obtain a visual region map; wherein, the visual region map includes... One visual area; The visual region map and the visual characteristic heatmap are processed into grayscale to obtain the visual region grayscale map. and visual characteristic thermal grayscale image ; against For each visual region, calculate the corresponding grayscale image. With grayscale image The average similarity is obtained. Average similarity; according to Average similarity is used to calculate the driver's regional visual characteristic value. : ； in, For the first The difference coefficient of each visual region For the first Average similarity among visual regions.

7. The assessment system for the impact of tunnel structure damage repair on traffic safety according to claim 6, characterized in that: The grayscale image corresponding to each visual region is calculated using the following formula. With grayscale image Average similarity: ； in, Represents pixels in the target visual region Corresponding grayscale image With grayscale image Average similarity; Indicates the brightness contrast value; ; Representing an image The average value of a pixel; For image The average value of a pixel; It is a constant; Indicates the contrast ratio; ; Representing an image Standard deviation of pixels; For image Standard deviation of pixels; It is a constant; Indicates the structural comparison value; ; Representing an image With images covariance; It is a constant.

Images