Road potential safety hazard monitoring system based on municipal engineering

Abstract

The invention relates to the technical field of image enhancement, and provides a road potential safety hazard monitoring system based on municipal engineering, and the system comprises an image obtaining module which obtains a municipal road image; the image analysis module obtains the area where the municipal road is located according to the gray level distribution; obtaining a sliding detection window of each pixel point in an area where the municipal road is located; acquiring dark particle pixel points in the sliding detection window according to the gray level distribution and the edge features; obtaining the discrete distribution degree of the dark particles according to the position distribution of the dark particle pixel points; obtaining the maximum width moment of the closed edge in the sliding detection window and the defect coefficient of each side length according to the distance between the edge lines; the municipal road surface defect degree is obtained; and the image enhancement module is used for improving a retinex algorithm by combining the municipal road surface defect degree and the discrete distribution degree so as to realize municipal road image enhancement. The invention aims to improve the definition of municipal road image enhancement, so that the details of the municipal road image are more prominent.

Description

Technical Field

[0001] The present invention relates to the field of image enhancement, and particularly to a road safety hazard monitoring system based on municipal engineering. Background Art

[0002] With the acceleration of urbanization, as the core carrier of the urban transportation network, the safety and reliability of municipal roads are directly related to public life and property and urban operation efficiency. According to the national road traffic safety plan, China needs to focus on improving the intrinsic safety level of road infrastructure and achieving active risk prevention and control and precise governance through scientific and technological means. However, traditional monitoring technologies are limited by environmental complexity and equipment performance, and it is difficult to capture road cracks, collapses, blurred signs and markings and other hazards in real time. Especially under low visibility conditions such as rain, fog and night, the quality of image acquisition drops significantly, resulting in low manual interpretation efficiency and high misjudgment rate.

[0003] Currently, road hazard detection systems based on computer vision generally rely on image enhancement technologies (such as the Retinex algorithm) to improve image quality. However, the single-scale Retinex (SSR) algorithm has the limitation of being sensitive to parameters. The selection of the σ value in the algorithm directly affects the enhancement effect, but traditional methods rely on empirical values or global fixed parameters and are difficult to meet the different requirements of different scenarios (such as crack edges and road surface reflection areas).

[0004] In summary, the present invention proposes a road safety hazard monitoring system based on municipal engineering. By acquiring municipal road images, a normalized municipal road safety coefficient is obtained according to the distribution of defective edge pixels of the municipal road, and the product of the normalized municipal road safety coefficient and the initial input value is used as the sigma value in the retinex algorithm to improve the clarity of image enhancement. Summary of the Invention

[0005] In order to solve the above technical problems, the present invention provides a road safety hazard monitoring system based on municipal engineering to solve the existing problems.

[0006] The technical solution adopted by the road safety hazard monitoring system based on municipal engineering of the present invention is as follows: An embodiment of the present invention provides a road safety hazard monitoring system based on municipal engineering, and the system includes the following modules: Image acquisition module: Acquire municipal road images; Image analysis module: Obtain the area where the municipal road is located in the municipal road image; perform edge detection on the area where the municipal road is located to obtain a set of initial pixel points of the closed edge; obtain the sliding detection window of each pixel point within the area where the municipal road is located; correct the sliding detection window of each pixel point according to the set of closed edge pixel points to obtain a new detection window; take each pixel point with a gray value less than the threshold in the new detection window as a dark particle pixel point; obtain the discrete distribution degree of dark particles according to the position distribution of the dark particle pixel points; obtain the maximum width distance of the closed edge within the new detection window; obtain the first side length defect coefficient, the second side length defect coefficient, and the third side length defect coefficient within the new detection window according to the maximum width moment of the closed edge and the gray value of the edge pixel points; obtain the triangular model of the surface defect of the municipal road in the new detection window according to the first side length defect coefficient, the second side length defect coefficient, and the third side length defect coefficient within the new detection window; obtain the degree of the surface defect of the municipal road in the new detection window according to the angle of the triangular model of the surface defect of the municipal road in the new detection window. Image enhancement module: Combine the degree of the surface defect of the municipal road in the new detection window and the discrete distribution degree of dark particles to obtain the discrimination degree of the municipal road defect in the new detection window; obtain the safety coefficient of the municipal road in the new detection window according to the discrimination degree of the municipal road defect in the new detection window; normalize the safety coefficient of the municipal road to obtain the normalized safety coefficient of the municipal road; use the normalized safety coefficient of the municipal road to correct the sigma value in the retinex algorithm to complete the enhancement of the municipal road image.

[0007] Preferably, the step of correcting the sliding detection window of each pixel point according to the set of closed edge pixel points to obtain a new detection window includes: For the sliding detection window of each pixel point, if there is a pixel point in the set of closed edge pixel points within the sliding detection window, obtain the direction from the central pixel point of the sliding detection window to the pixel point, and expand the sliding detection window along the direction until the entire closed edge is included in the sliding detection window, and take the expanded sliding detection window as the new detection window.

[0008] Preferably, the step of obtaining the discrete distribution degree of dark particles according to the position distribution of the dark particle pixel points includes: Calculate the Euclidean distance between each dark particle and the central pixel point of the sliding detection window, and take the mean value of all the Euclidean distances as the discrete distribution degree of dark particles.

[0009] Preferably, the step of obtaining the maximum width distance of the closed edge within the new detection window includes: For each closed edge, calculate the Euclidean distance between any two edge pixel points on the closed edge, and take the maximum value of the Euclidean distances as the maximum width distance of the closed edge.

[0010] Preferably, obtaining the first side length defect coefficient, the second side length defect coefficient, and the third side length defect coefficient within the new detection window according to the maximum width of the closed edge and the gray values of the edge pixel points includes: Obtain two pixel points that constitute the maximum width of the closed edge, respectively denoted as the first pixel point and the second pixel point. Obtain the pixel point with the minimum gray value on the closed edge, denoted as the third pixel point. Take the difference between the gray values of the first pixel point and the third pixel point, the difference between the gray values of the second pixel point and the third pixel point, and the Euclidean distance between the first pixel point and the second pixel point as the first side length defect coefficient, the second side length defect coefficient, and the third side length defect coefficient within the new detection window.

[0011] Preferably, obtaining the triangular model of the surface defect of the municipal road for the new detection window according to the first side length defect coefficient, the second side length defect coefficient, and the third side length defect coefficient within the new detection window includes: Take the first pixel point, the second pixel point, and the third pixel point as the three vertices of the triangular model of the surface defect of the municipal road, and take the first side length defect coefficient, the second side length defect coefficient, and the third side length defect coefficient as the side lengths between the first pixel point and the third pixel point, the second pixel point and the third pixel point, and the first pixel point and the second pixel point in the triangular model of the surface defect of the municipal road, respectively.

[0012] Preferably, obtaining the degree of the surface defect of the municipal road for the new detection window according to the angles of the triangular model of the surface defect of the municipal road for the new detection window, the expression is: In the formula, is the degree of the surface defect of the municipal road, is the side length defect coefficient of side BC, is the side length defect coefficient of side AC, is the side length defect coefficient of side AB, is the inverse cosine trigonometric function.

[0013] Preferably, obtaining the discrimination degree of the municipal road defect for the new detection window by combining the degree of the surface defect of the municipal road for the new detection window and the discrete distribution degree of the dark particles includes: For the new detection window, take the ratio of the degree of the surface defect of the municipal road to the discrete distribution degree of the dark particles as the discrimination degree of the municipal road defect for the new detection window.

[0014] Preferably, obtaining the safety coefficient of the municipal road for the new detection window according to the discrimination degree of the municipal road defect for the new detection window includes: For the new detection window, take the reciprocal of the discrimination degree of the municipal road defect as the safety coefficient of the municipal road for the new detection window.

[0015] Preferably, the enhancement of the municipal road image by modifying the sigma value in the retinex algorithm using the normalized municipal road safety factor includes: Set the initial input value, and use the product of the initial input value and the normalized municipal road safety factor as the sigma value in the retinex algorithm.

[0016] The present invention has at least the following beneficial effects: The present invention mainly enhances the municipal road image through the gray-scale features of the municipal road image and the edge contour features of the defects, obtains the normalized municipal road safety factor of each new detection window, replaces the sigma value in the retinex algorithm, realizes the enhancement of the municipal road image, and improves the authenticity and naturalness of the municipal road image enhancement. The present invention combines the discrimination degree of municipal road defects and the normalized municipal road safety factor of each new detection window for comprehensive analysis, and enhances the image of the municipal road in regions, making the detailed information of the municipal road image more prominent; Furthermore, the present invention first obtains the set of initial pixel points of the closed edge, sets a sliding detection window. When the pixel points in the sliding detection window coincide with the pixel points in the set of initial pixel points of the closed edge, the sliding detection window is adjusted to include the entire closed edge. By analyzing the width of the closed edge and the gray scale of the pixel points, the normalized municipal road safety factor of the new detection window is obtained, realizing a road safety hazard monitoring system based on municipal engineering, and solving the problem that due to the defects of the municipal road, the image is blurred due to dust accumulation, resulting in an unsatisfactory image enhancement effect. The present invention has the beneficial effects of real and natural image enhancement and prominent details. Description of the Drawings

[0017] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0018] Figure 1 It is a module flowchart of a road safety hazard monitoring system based on municipal engineering provided by an embodiment of the present invention; Figure 2 It is a schematic diagram of a triangular model of surface defects of a municipal road. Detailed Embodiments

[0019] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation method, structure, features and effects of a road safety hazard monitoring system based on municipal engineering proposed by the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0020] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0021] The following is a detailed description of a specific scheme of a road safety hazard monitoring system based on municipal engineering provided by the present invention in conjunction with the accompanying drawings.

[0022] See also Figure 1 , which shows a module flow chart of a road safety hazard monitoring system based on municipal engineering provided by an embodiment of the present invention, the system includes the following modules: Image acquisition module: Acquire municipal road images through the image acquisition device and perform preprocessing.

[0023] Specifically, firstly, this embodiment will use a camera to obtain an RGB image of a municipal road as a data source for municipal road image enhancement. It should be noted that there are many methods for obtaining municipal road images. The specific image acquisition method can be implemented by existing technologies and is not within the scope of protection of this embodiment, so no detailed description will be given. Then, the municipal road image is converted into a grayscale image, and the specific method is the average method. Secondly, the municipal road grayscale image is denoised using a guided filtering denoising algorithm to remove noise interference. Since the average method and the guided filtering denoising algorithm are both existing well-known technologies, they will not be described in detail here; At this point, the denoised grayscale image of the municipal road can be obtained according to the above method of this embodiment, which serves as the data basis for subsequent municipal road image enhancement.

[0024] Image analysis module: The closed edges in the municipal road image are obtained based on the defect characteristics of the municipal road, and the normalized municipal road safety factor is obtained based on the width and grayscale distribution of the closed edges.

[0025] Specifically, in this embodiment, the area where the municipal road is located is obtained according to the gray-scale distribution; the sliding detection window of each pixel point in the area where the municipal road is located is acquired; the dark particle pixel points in the sliding detection window are obtained according to the gray-scale distribution and edge features; the discrete distribution degree of the dark particles is obtained according to the position distribution of the dark particle pixel points; the maximum width moment of the closed edge and the side length defect coefficient in the sliding detection window are obtained according to the distance between the edge lines; thereby, the triangular model of the surface defects of the municipal road is obtained to calculate the degree of the surface defects of the municipal road, and the sigma parameter in the retinex algorithm is improved by combining the degree of the surface defects of the municipal road and the discrete distribution degree, so as to realize the enhancement of the municipal road image. The specific construction process of the normalized safety coefficient of the new detection window is as follows: The situations that may cause potential safety hazards considered in the embodiment of the present invention for the municipal road are: cracks, depressions, and fractures. Among them, cracks are generally strip-shaped. Cracks formed by non-human factors are mostly irregular strip-shaped lines, and may also have characteristics such as bifurcations or locally dense distributions; scratch-type cracks caused by human factors such as working equipment, handling and installation of the municipal road are relatively regular, presenting strip-shaped cracks with sharp sides and a wider middle. The situation of depression is the local indentation of the surface of the municipal road caused by external factors, which may present a round depression (the depression area is relatively smooth) and an irregular depression (the middle area of the depression is uneven) according to different external factors. The fracture situation is more serious, usually with lateral offsets on some areas of the municipal road and large-area vacancies in areas that should be continuous originally. Analyses are made based on the above several defect situations of the municipal road.

[0026] First, the Otsu threshold segmentation method is used for the gray-scale image of the municipal road obtained by the image acquisition module to obtain the area of the municipal road in the image. The Otsu threshold segmentation method is a well-known existing technology, and this embodiment will not elaborate too much here. A binary image with the area of the municipal road as the foreground and other parts as the background is obtained. According to the pixel point positions in the area of the municipal road in this image, position mapping is performed on the original gray-scale image, and the area of the municipal road is marked as area A, and other positions are marked as area B. Finally, the area A where the municipal road is located in the original gray-scale image is obtained.

[0027] Then, canny edge detection is performed on the municipal road area A. The canny edge detection is a well-known existing technology, and this embodiment will not elaborate too much here. After obtaining the edges, a closedness detection is performed on each edge. In this embodiment, the chain code method is adopted. First, any pixel point on the edge is used as the initial point, and its coordinates are recorded as a , and then other pixel points in each direction of this pixel point are searched. If the other pixel points are also edge pixel points, the two are connected, and it is used as the next initial point to continue traversing. When the final coordinates are the same as the initial point coordinates, it means that this edge line forms a closed edge. After finding the edge line that can form a closed edge, the initial point a of this edge line Record the initial pixel points of the closed edge into the set P, and classify the pixel points on the edge line that do not form a closed edge into the pixel point set Q. Then, create a sliding detection window centered on each pixel point in area A, and perform a sliding search on area A. In this embodiment, , the implementer can set it according to the actual situation.

[0028] Safety hazards in municipal roads are often due to various defects on the surface of municipal roads. In most cases, the impact of the defects is that local deformation or cracks occur on the surface of the municipal roads, usually presenting as deep closed areas. The greater the difference in gray value between the deeper part in the middle of the closed area and the edge, the greater the degree of the defect. If the closed area is wider, it indicates that the tightness of the area is smaller, and the possibility of the area being a defect is also greater.

[0029] The distribution position and brightness of light sources in the image of the municipal road surface are usually stable, and the brightness of the light spots generated by the influence of light sources on the municipal road surface is also relatively constant. When dust deposits on the smooth surface of the municipal road, only a small amount of dust will be on the surface of the municipal road, resulting in relatively discrete and less significant specular reflection effects. However, when there are defects in the municipal road, the defect area will have a weakened light reflection ability due to the deposition of dust. If the dust deposition amount in an area is small, most of the pixel points in that area have a strong light reflection ability, and a small number of pixel points have a weak light reflection ability because they are blocked by dust, and the overall area brightness is relatively high. If the dust deposition amount in an area is large, most of the pixel points are blocked by dust and have a weak light reflection ability, and the remaining small number of pixel points with a strong reflection ability result in a relatively dark overall area brightness. In summary, if there are defects in the municipal road, the larger the defect, the greater the local dust coverage, and the smaller the defect or no defect, the smaller the local dust coverage.

[0030] In response to the above analysis, when a pixel point in the set P of the initial pixel points of the closed edge appears in the sliding detection window, connect the center point of the sliding detection window with the pixel point in the set P that is closest to the center point of the sliding detection window. Denote the coordinates of the center point of the sliding detection window as ( , ) and the gray value of the center point as , to form the expansion direction of the sliding detection window. Based on this direction, expand the sliding detection window in this direction until the entire closed edge is surrounded by the sliding detection window. At this time, use this sliding detection window as the new detection window, and calculate the average gray value of all pixel points in the new detection window, and use this average gray value Threshold segmentation is performed on the pixel points within the new detection window using a threshold. Pixel points with a gray value greater than the threshold are classified as bright particles, and pixel points with a gray value lower than the threshold are classified as dark particles. That is, the pixel points covered by dust are dark particles. The formation of dark particles is due to the deposition of dust on the surface of the municipal road, resulting in a weakened sensitivity of some pixel points on the surface of the municipal road to the light source. Since the distribution of dust on the surface of the defect-free municipal road is discrete, the distribution of dark particles in the new detection window is also discrete. However, when there are defects on the surface of the municipal road, the dust deposition amount in the defect area increases, and the number of dark particles in the defect area also increases, showing an aggregated distribution with a small degree of dispersion. Record the coordinates of each dark particle pixel point ( , )(j = 1, 2,..., z), and calculate the distance between each dark particle pixel point and the coordinates of the center point of the sliding detection window. The specific expression is: In the formula, is the distance of the dark particle from the center point of the sliding detection window, is the abscissa of the center point of the sliding detection window, is the ordinate of the center point of the sliding detection window, is the abscissa of the dark particle, is the ordinate of the dark particle.

[0031] Then, the degree of discrete distribution of dark particle pixel points within the new detection window is obtained by the average distance method. The specific expression of the degree of discrete distribution is: In the formula, is the degree of discrete distribution of dark particle pixel points within the new detection window, is the distance of the dark particle from the center point of the sliding detection window, and z is the total number of dark particle pixel points in the new detection window.

[0032] The greater the degree of discrete distribution of dark particles, the lower the possibility that the area where the new detection window is located is a defect area of the municipal road. The smaller the degree of discrete distribution of dark particles, the higher the possibility that the area where the new detection window is located is a defect area of the municipal road.

[0033] Then, any two points on the closed edge line are connected to obtain the Euclidean distance (i = 1, 2,..., n) between the two points, and the maximum width distance of the closed edge is obtained among all the distances. The specific expression of the maximum width distance of the closed edge is: In the formula, is the maximum width of the closed edge of the closed edge line, and n is the total number of points on the edge line. is the maximum value function.

[0034] Then, the two pixel points that can form the maximum width of the closed edge are denoted as point A and point B, which are respectively denoted as the first pixel point and the second pixel point, and the gray values are respectively recorded as and . Then, the gray values of the pixel points on the closed edge are statistically analyzed, and the gray value of each point is recorded (n = 1, 2,..., m), and the minimum gray value of the pixel points on the closed edge is statistically obtained, denoted as . The pixel points with the gray value of are marked as point C, denoted as the third pixel point.

[0035] Obtain the three pixel points A, B, and C, and construct a triangular model of the surface defect of the municipal road based on the gray value differences of the three pixel points A, B, and C. Specifically, obtain and and The differences are used as the BC side length defect coefficient and the AC side length defect coefficient of the triangular model of the surface defect of the municipal road, and are respectively denoted as the second side length defect coefficient and the first side length defect coefficient. The specific expressions of the BC side length defect coefficient and the AC side length defect coefficient are: In the formula, is the BC side length defect coefficient, is the gray value of pixel point A, is the minimum gray value of the pixel points on the closed edge, is the AC side length defect coefficient, is the gray value of pixel point B.

[0036] Finally, let be the AB side length defect coefficient of the triangular model of the surface defect of the municipal road, denoted as the third side length defect coefficient. The schematic diagram of the triangular model of the surface defect of the municipal road is as Figure 2 shown.

[0037] Since the side length defect coefficients a, b, and c of the triangular model of the surface defect of the municipal road are known, the degree of the surface defect of the municipal road can be calculated. The specific expression of the degree of the surface defect of the municipal road is: In the formula, k is the degree of the surface defect of the municipal road, representing the included angle between the BC side and the AC side, is the BC side length defect coefficient, is the AC side length defect coefficient, is the defect coefficient of side AB, is the inverse cosine trigonometric function.

[0038] The larger the value of k, the larger the area covered by the closed edge, the larger the range of municipal road defects, and the larger the dust coverage within the municipal road defects.

[0039] According to the above, the discrimination degree of municipal road defects within the detection window can be obtained. The specific expression of the discrimination degree of municipal road defects is: In the formula, HS is the discrimination degree of municipal road defects in the new detection window, k is the cosine value of the angle θ of the approximate triangle formed by the closed area, and q is the degree of discrete distribution of dark particles. Since when there are defects in the municipal road, the larger the defect, the greater the degree of dust deposition in the defect, and the smaller the defect or no defect, the smaller the degree of local dust deposition. When the value of HS is larger, it indicates that the distance between the closed areas within the detection window is larger, the dust distribution is concentrated, and it is more likely to be the municipal road defect area; when the value of HS is smaller, it indicates that the distance between the closed areas within the detection window is smaller, the dust distribution is discrete, and it is less likely to be the municipal road defect area.

[0040] Thus, the discrimination degree HS of municipal road defects is obtained.

[0041] Image enhancement module: Locally correct the sigma value of the retinex algorithm to achieve image enhancement of the municipal road.

[0042] The embodiment of the present invention is image enhancement based on the retinex theory, mainly using the single-scale SSR enhancement algorithm in retinex. Among them, the smaller the value of sigma in the single-scale SSR algorithm, the better the enhancement effect on image details. Although a smaller value of sigma may cause color distortion or halos in the image, the ultimate goal of the present invention is to make the defects in the municipal road more obvious in the image, and the small degree of distortion and halos in the local area are acceptable. According to the indicators constructed above, the value of sigma in the SSR algorithm is restricted and adjusted to construct the municipal road safety factor. The specific expression of the municipal road safety factor is: In the formula, is the municipal road safety factor of the new detection window, is the discrimination degree of municipal road defects in the new detection window, is the natural constant. To ensure that the numerator is not zero, in this embodiment , and the implementer can set it according to the actual situation.

[0043] When the safety factor of the municipal road is smaller, it indicates a greater possibility of defects in the area. The value of sigma should be correspondingly taken as a smaller value to enhance the area where defects may exist. It should be noted that the value range of sigma in the single-scale SSR is [80, 100]. Therefore, it is necessary to adjust the parameters for normalization to ensure that the value range of the adjusted sigma value is still [80, 100].

[0044] Construct the normalized safety factor of the municipal road. The specific expression of the normalized safety factor of the municipal road is as follows: In the formula, is the normalized safety factor of the municipal road, is the safety factor of the municipal road in the new detection window, and d are both normalized adjustment parameters, is the function to take the minimum value, is the function to take the maximum value. is to perform normalization processing on so that , to ensure the feasibility of the retinex algorithm. In this embodiment, , d = 0.8. The implementer can set it according to the actual situation. Set the initial input value to 100, is the value of sigma finally input into the SSR algorithm. When is smaller, it indicates a greater probability that the area is a defect of the municipal road. Therefore, the value of is smaller.

[0045] Finally, for the area where each new detection window is located, take the calculated value as the input parameter of the retinex algorithm. For other areas, the sigma value of the retinex algorithm is the initial input value . Through the above steps, the enhancement ability of the local area details of the municipal road image is stronger. In the finally output image, the areas with defects in the municipal road are more prominent, which allows the staff to discover and handle relevant potential safety hazards earlier, greatly improving the safety guarantee.

[0046] To sum up, the embodiment of the present invention solves the problem that due to defects in the municipal road, the image enhancement effect is not ideal due to image blurring caused by dust accumulation. By using the retinex algorithm and combining the edge features of the municipal road defects, the authenticity of the municipal road image enhancement is improved, and the local detail information of the image is more prominent.

[0047] It should be noted that: The above order of the embodiments of the present invention is only for description and does not represent the superiority or inferiority of the embodiments. In addition, the specific embodiments of this specification have been described. Moreover, the processes depicted in the drawings do not necessarily require the specific order or consecutive order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0048] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments.

[0049] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A road safety hazard monitoring system based on municipal engineering, characterized in that, The system includes the following modules: Image acquisition module: Acquire municipal road images; Image analysis module: Obtain the area where the municipal road is located in the municipal road image; Perform edge detection on the area where the municipal road is located to obtain a set of initial pixel points of the closed edge; Obtain the sliding detection window of each pixel point within the area where the municipal road is located; Modify the sliding detection window of each pixel point according to the set of closed edge pixel points to obtain a new detection window; Take each pixel point with a gray value less than the threshold in the new detection window as a dark particle pixel point; Obtain the discrete distribution degree of dark particles according to the position distribution of dark particle pixel points; Obtain the maximum width of the closed edge within the new detection window; Obtain the first side length defect coefficient, the second side length defect coefficient, and the third side length defect coefficient within the new detection window according to the maximum width of the closed edge and the gray values of the edge pixel points; Obtain the triangular model of the surface defect of the municipal road in the new detection window according to the first side length defect coefficient, the second side length defect coefficient, and the third side length defect coefficient within the new detection window; Obtain the degree of surface defect of the municipal road in the new detection window according to the angle of the triangular model of the surface defect of the municipal road in the new detection window; Image enhancement module: Obtain the discrimination degree of the municipal road defect in the new detection window by combining the degree of surface defect of the municipal road in the new detection window and the discrete distribution degree of dark particles; Obtain the safety coefficient of the municipal road in the new detection window according to the discrimination degree of the municipal road defect in the new detection window; Normalize the safety coefficient of the municipal road to obtain the normalized safety coefficient of the municipal road; Modify the sigma value in the retinex algorithm using the normalized safety coefficient of the municipal road to complete the enhancement of the municipal road image.

2. The road safety hazard monitoring system based on municipal engineering according to claim 1, characterized in that, The step of modifying the sliding detection window of each pixel point according to the set of closed edge pixel points to obtain a new detection window includes: For the sliding detection window of each pixel point, if there is a pixel point in the set of closed edge pixel points within the sliding detection window, obtain the direction from the central pixel point of the sliding detection window to the pixel point, and expand the sliding detection window along the direction until the entire closed edge is included in the sliding detection window, and take the expanded sliding detection window as the new detection window.

3. A road safety hazard monitoring system based on municipal engineering according to claim 1, characterized in that, The step of obtaining the discrete distribution degree of dark particles according to the position distribution of dark particle pixel points includes: Calculate the Euclidean distance between each dark particle and the central pixel point of the sliding detection window, and take the mean of all the Euclidean distances as the discrete distribution degree of dark particles.

4. A road safety hazard monitoring system based on municipal engineering according to claim 1, characterized in that, The step of obtaining the maximum width of the closed edge within the new detection window includes: For each closed edge, calculate the Euclidean distance between any two edge pixel points on the closed edge, and take the maximum value of the Euclidean distances as the maximum width of the closed edge.

5. A road safety hazard monitoring system based on municipal engineering according to claim 1, characterized in that, The step of obtaining the first side length defect coefficient, the second side length defect coefficient, and the third side length defect coefficient within the new detection window according to the maximum width of the closed edge and the gray values of the edge pixel points includes: Obtain two pixel points that constitute the maximum width distance of the closed edge, denoted as the first pixel point and the second pixel point respectively. Obtain the pixel point with the minimum gray value on the closed edge, denoted as the third pixel point. Take the gray value difference between the first pixel point and the third pixel point, the gray value difference between the second pixel point and the third pixel point, and the Euclidean distance between the first pixel point and the second pixel point as the first side length defect coefficient, the second side length defect coefficient, and the third side length defect coefficient within the new detection window.

6. The road safety hazard monitoring system based on municipal engineering according to claim 5, characterized in that, The obtaining of the triangular model of the surface defect of the municipal road for the new detection window based on the first side length defect coefficient, the second side length defect coefficient, and the third side length defect coefficient within the new detection window includes: Take the first pixel point, the second pixel point, and the third pixel point as the three vertices of the triangular model of the surface defect of the municipal road respectively. Take the first side length defect coefficient, the second side length defect coefficient, and the third side length defect coefficient as the side lengths between the first pixel point and the third pixel point, the second pixel point and the third pixel point, and the first pixel point and the second pixel point in the triangular model of the surface defect of the municipal road respectively.

7. A road safety hazard monitoring system based on municipal engineering according to claim 1, characterized in that, The degree of surface defects of the municipal road in the new detection window is obtained according to the angle of the triangular model of surface defects of the municipal road in the new detection window, and the expression is: Wherein, is the degree of surface defects of the municipal road, is the defect coefficient of the BC side length, is the defect coefficient of the AC side length, is the defect coefficient of the AB side length, is the inverse cosine trigonometric function.

8. A road safety hazard monitoring system based on municipal engineering according to claim 1, characterized in that, The obtaining of the discrimination degree of the municipal road defect for the new detection window by combining the surface defect degree of the municipal road in the new detection window and the discrete distribution degree of the dark particles includes: For the new detection window, take the ratio of the surface defect degree of the municipal road to the discrete distribution degree of the dark particles as the discrimination degree of the municipal road defect for the new detection window.

9. The road safety hazard monitoring system based on municipal engineering according to claim 1, characterized in that, The obtaining of the safety coefficient of the municipal road for the new detection window based on the discrimination degree of the municipal road defect for the new detection window includes: For the new detection window, take the reciprocal of the discrimination degree of the municipal road defect as the safety coefficient of the municipal road for the new detection window.

10. A road safety hazard monitoring system based on municipal engineering according to claim 1, characterized in that, The enhancing of the municipal road image by using the normalized safety coefficient of the municipal road to correct the sigma value in the retinex algorithm includes: Set the initial input value, and take the product of the initial input value and the normalized safety coefficient of the municipal road as the sigma value in the retinex algorithm.

Images