Road construction quality detection method

Abstract

The invention discloses a highway construction quality detection method. The method comprises the following steps: obtaining a road image subjected to image preprocessing, and determining each initial recess sub-region in the road image; according to the gray value of each pixel point in each concave sub-region and the number of the pixel points, determining a morphological characteristic value of the concave sub-region; according to the morphological characteristic value, judging whether a first concave sub-region exists or not; if yes, determining a spatial distribution characteristic value of the first concave sub-region according to the form characteristic value and the position of each pixel point; according to the spatial distribution characteristic value, judging whether a second concave sub-region exists or not; if yes, the number of pixel points of the second sunken sub-area is obtained, and whether sunken parts are caused by overweight of the vehicle or not is judged; and removing the second concave sub-region to form a third concave sub-region, and analyzing the distribution characteristics of the third concave sub-region. The method can effectively identify and eliminate the road surface sunken area caused by overweight of the vehicle, and improves the accuracy and practicability of the detection result.

Description

Technical Field

[0001] The invention relates to the technical field of image processing, and in particular to a method for detecting highway construction quality. Background Art

[0002] Highway construction takes a long time, and quality problems are prone to occur during the construction process. The quality of highway construction affects the performance and life of the highway, and even for highways with lower quality, it may bring great safety hazards. Highways usually use asphalt concrete pavement, but during construction, unreasonable docking of the two constructions may cause longitudinal and transverse gaps in the asphalt highway pavement. Pavement construction needs to properly deal with the gap problem of asphalt highway pavement, provided that these gaps are identified first.

[0003] With the rapid development of computer vision, the use of image processing technology to identify gaps in asphalt road surfaces has become mainstream. Image segmentation algorithms, such as the segmentation algorithm based on region growing, are used. This algorithm is less sensitive to noise and can automatically identify various objects. However, the complexity of asphalt road surfaces makes the results of region growing more complicated, which in turn affects the subsequent quality inspection results.

[0004] In addition, the inaccurate connection between the two constructions will not only cause gaps in the asphalt road surface, but also cause bulges and depressions on the road surface. The traditional region growing algorithm is calculated based on the gray value of each pixel in the image. There are bulges and depressions in the constructed asphalt road surface, and the gray value changes in these areas will directly affect the region growing algorithm, which is easy to misclassify them as crack areas, resulting in inaccurate detection results.

[0005] For this purpose, a highway construction quality detection method based on computer vision is proposed (patent name: Highway construction quality detection method based on computer vision, patent number: CN202410046051.1). This method calculates the gap characteristic coefficient and edge clutter index of each pixel point, and uses the region growing algorithm to perform region growth based on the similarity of the gap characteristic values ​​between adjacent pixels, thereby obtaining a more accurate asphalt highway pavement gap area. However, in actual use, roads often sag due to overweight vehicles, resulting in low-lying areas. These low-lying areas may be mistakenly included in the construction quality detection content, resulting in an imbalance in the detection results and inconvenience in use. Summary of the invention

[0006] The purpose of the present invention is to provide a highway construction quality detection method, which can effectively identify and eliminate road surface depression areas caused by vehicle overweight, and improve the accuracy and practicality of the detection results.

[0007] To achieve the above object, the present invention provides the following technical solution: a method for detecting the quality of highway construction, comprising the following steps:

[0008] S1, obtaining a highway image that has undergone image preprocessing, including the steps of denoising, contrast enhancement, and distortion correction, and determining each initial concave sub-region in the highway image according to the highway image;

[0009] S2, determining the morphological feature value of each initial concave sub-region according to the gray value of each pixel point in each initial concave sub-region and the number of pixels points in each initial concave sub-region;

[0010] S3, judging whether there is a first concave sub-region in each of the initial concave sub-regions according to the morphological feature values ​​of each of the initial concave sub-regions;

[0011] S31, if it exists, determining the spatial distribution characteristic value of each first concave sub-region according to the morphological characteristic value of each initial concave sub-region, each pixel point in each first concave sub-region and the position of each pixel point;

[0012] S4, judging whether there is a second concave sub-region in each of the first concave sub-regions according to the spatial distribution characteristic values ​​of each of the first concave sub-regions;

[0013] S41, if yes, obtaining the number of pixels in the second concave sub-region, and judging whether there is a concave caused by the overweight vehicle in the highway image according to the number of pixels in the second concave sub-region;

[0014] S5, removing the second concave sub-region from the first concave sub-region to form a third concave sub-region;

[0015] S6. Detect and analyze the distribution characteristics of the third concave sub-region.

[0016] Preferably, the calculation of the morphological feature value comprises the following steps:

[0017] Calculate the average gray value (G m );

[0018] Calculate the edge density (E) of each initial concave sub-region d );

[0019] Calculate the shape complexity (S) of each initial concave sub-region c );

[0020] Calculate the area (A) of each initial concave sub-region;

[0021] The calculation formula of the morphological characteristic value is: F(t) = αG m +βE d+γS c +δA;

[0022] Among them, α, β, γ, and δ are weight coefficients, which are adjusted according to the specific application scenario.

[0023] Preferably, the calculation formula of the average gray value is: Where N is the total number of pixels in the concave sub-region, G i is the gray value of the i-th pixel;

[0024] The edge density is calculated as follows: E is the total number of edge pixels in the concave sub-region, and A is the area of ​​the concave sub-region;

[0025] The calculation formula of the shape complexity is: Where P is the perimeter of the concave sub-region, and A is the area of ​​the concave sub-region;

[0026] The area of ​​the concave sub-region is calculated as follows: Wherein, N is the total number of pixels in the concave sub-region.

[0027] Preferably, the weight coefficients α, β, γ, and δ are calculated as follows:

[0028]

[0029] Among them, μ G , μ E , μ S , μ A are the average gray value, edge density, shape complexity and area respectively, and N, M, P, Q are the total number of samples of the corresponding eigenvalues ​​respectively.

[0030] Preferably, determining the spatial distribution characteristic value of each first depressed sub-region according to the morphological characteristic value of each initial depressed sub-region, each pixel point in each first depressed sub-region and the position of each pixel point includes:

[0031] Acquire adjacent regions of each first concave sub-region;

[0032] Determine the distance characteristics of each first concave sub-region according to the edge pixel positions of each first concave sub-region and its adjacent region, including the minimum distance, the maximum distance, the average distance and the standard deviation;

[0033] Further refining the morphological feature values ​​of each first concave sub-region according to the morphological feature values ​​of each initial concave sub-region;

[0034] Determining the aspect ratio of each first recessed sub-region according to the width and height of each first recessed sub-region and its adjacent region;

[0035] Fitting all pixel points in each first concave sub-region and its adjacent region with a straight line equation to obtain the inclination angle of each first concave sub-region and its adjacent region;

[0036] The spatial distribution characteristic value of each first recessed sub-region is determined according to the distance characteristic, the morphological characteristic value, the aspect ratio and the inclination angle of each first recessed sub-region.

[0037] Preferably, the calculation of the spatial distribution characteristic value comprises the following steps:

[0038] B1. Calculate the minimum distance between the first concave sub-region and the adjacent region, and the calculation formula is: D min =min({d i}), where d i is the distance between the edge pixels of the first concave sub-region and the edge pixels of the adjacent region;

[0039] B2. Calculate the maximum distance between the first concave sub-region and the adjacent region, and the calculation formula is: D max =max({d i}), where d i is the distance between the edge pixels of the first concave sub-region and the edge pixels of the adjacent region;

[0040] B3. Calculate the average distance between the first concave sub-region and the adjacent region, and the calculation formula is:

[0041] where d i is the distance between the edge pixel of the first concave sub-region and the edge pixel of the adjacent region, and n is the total number of distances;

[0042] B4. Calculate the standard deviation of the distance between the first concave sub-region and the adjacent regions, and the calculation formula is: where d i is the distance between the edge pixel of the first concave sub-region and the edge pixel of the adjacent region, D mean is the average distance, n is the total number of distances;

[0043] B5. Calculate the aspect ratio of the first concave sub-region using the following calculation formula: Wherein, W is the width of the first recessed sub-region, and H is the height of the first recessed sub-region;

[0044] B6. Calculate the inclination angle of the first concave sub-region, and the calculation formula is: Wherein Δy is the longitudinal change of the pixel point in the first concave sub-area, and Δx is the lateral change;

[0045] B7. The morphological characteristic value F(t) is used as one of the input parameters to calculate the comprehensive spatial distribution characteristic value. The calculation formula is: S d =F(t)+D min +D max +D mean +D σ +R aspect +θ, where F(t) is the calculated morphological feature value.

[0046] Preferably, judging whether there is a second depressed sub-region in each first depressed sub-region according to the spatial distribution characteristic value of each first depressed sub-region comprises the following steps:

[0047] The spatial distribution characteristic value S of the first concave sub-region d Compare with a preset threshold;

[0048] If S d If the threshold is exceeded, the first concave sub-region is marked as a potential second concave sub-region;

[0049] The first depressed sub-region marked as a potential second depressed sub-region is further analyzed to determine its specific position and range.

[0050] Preferably, if there is a second concave sub-region, the number of pixels in the second concave sub-region is obtained, and according to the number of pixels in the second concave sub-region, it is determined whether there is a concave caused by an overweight vehicle in the highway image, including the following steps:

[0051] Calculating the number of pixels in the second concave sub-region;

[0052] Compare the calculated result with a preset vehicle overweight depression characteristic value;

[0053] If the number of pixels exceeds the preset value, it is determined that the area may be caused by overweight vehicles;

[0054] The judgment results are marked on the highway image for subsequent processing and maintenance.

[0055] Preferably, removing the second recessed sub-region in the first recessed sub-region to form a third recessed sub-region comprises the following steps:

[0056] Mark and remove the second concave sub-region in the first concave sub-region;

[0057] Recalculating the remaining concave area to form a third concave sub-area;

[0058] The highway image is updated to reflect the specific location and range of the third concave sub-region.

[0059] Preferably, analyzing the distribution characteristics of the third recessed sub-region comprises the following steps:

[0060] Calculating the distribution characteristics of the third concave sub-region, including area, depth and shape complexity;

[0061] Compare the calculated results with the standard distribution characteristics to evaluate construction quality issues;

[0062] Generate a detailed inspection report with specific information and maintenance recommendations for the third depression sub-area;

[0063] The test report is stored in the database for subsequent query and analysis.

[0064] Compared with the prior art, the present invention has the following beneficial effects:

[0065] 1. The present invention improves image quality and ensures the accuracy of subsequent detection by setting detailed image preprocessing steps;

[0066] 2. The present invention can effectively identify and segment the initial concave sub-regions in the highway image, reduce the influence of noise, and enhance the accuracy of identification;

[0067] 3. By calculating the morphological characteristic values ​​and spatial distribution characteristic values ​​of the initial concave sub-region, the present invention can accurately determine the position and range of the concave region, avoiding the detection error caused by image complexity in the traditional method;

[0068] 4. The present invention can eliminate the second concave sub-region caused by vehicle overweight and form a third concave sub-region, thereby improving the reliability and practicality of the detection result;

[0069] 5. By conducting a detailed analysis of the distribution characteristics of the third depressed sub-area, the present invention can generate a comprehensive highway construction quality inspection report to provide a scientific basis for highway maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 This is a flow chart of the highway construction quality detection method of the present invention;

[0071] Figure 2 This is a flow chart of the calculation of morphological characteristic values ​​of the present invention;

[0072] Figure 3 This is a flow chart of calculating the spatial distribution characteristic value of the present invention;

[0073] Figure 4 This is a flow chart of the calculation of the comprehensive spatial distribution characteristic value of the present invention;

[0074] Figure 5 A flow chart of the present invention for determining whether a second concave sub-region exists;

[0075] Figure 6 This is a flow chart of judging vehicle overweight depression according to the present invention;

[0076] Figure 7 A flow chart for forming a third recessed sub-region for the present invention;

[0077] Figure 8 Analyze the third recessed sub-region flow chart for the present invention. DETAILED DESCRIPTION

[0078] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0079] See also Figures 1 to 8 The present invention provides a technical solution: The present invention discloses a highway construction quality detection method, which aims to improve the detection accuracy and reliability of highway construction quality, and specifically includes the following steps:

[0080] S1. Obtaining a highway image after image preprocessing:

[0081] First, the highway image is processed using image preprocessing technology. This includes the following steps:

[0082] Denoising: Use Gaussian filtering and other methods to remove noise from the image and improve image clarity.

[0083] Enhance contrast: Enhance the contrast of the image through techniques such as histogram equalization to make the concave areas more obvious.

[0084] Correct distortion: Use geometric correction algorithms to correct distortion in images, making the actual size and shape in the image more accurate.

[0085] After preprocessing, the initial sunken sub-regions of the highway are extracted from the image. These regions are determined based on the gray value changes in the image and the edge detection algorithm.

[0086] S2. Determine the morphological characteristic values ​​of each initial concave sub-region:

[0087] For each initial concave sub-region, the morphological feature values ​​of the region are calculated based on the grayscale value and number of pixels in it. These morphological feature values ​​include:

[0088] Mean gray value (Gm): This value reflects the overall brightness level of the concave area.

[0089] Edge density (Ed): indicates the density of edge pixels in the concave area.

[0090] Shape complexity (Sc): It reflects the shape complexity of the concave area by calculating the ratio of the perimeter to the area of ​​the concave area.

[0091] Area (A): The total number of pixels within the concave area.

[0092] S3. Determine whether there is a first concave sub-region:

[0093] According to the morphological feature values ​​of each initial concave sub-region, it is determined whether there is a first concave sub-region in these regions. The first concave sub-region refers to a region with significant concave features, and its morphological feature value exceeds a preset threshold.

[0094] S31, determining the spatial distribution characteristic value of the first concave sub-region:

[0095] If there is a first concave sub-region, the position and morphological feature value of each pixel in these regions are further analyzed to determine their spatial distribution feature values. These feature values ​​include:

[0096] Minimum distance (Dmin): the minimum distance between the first concave sub-region and the adjacent region.

[0097] Maximum distance (Dmax): the maximum distance between the first concave sub-region and the adjacent region.

[0098] Average distance (Dmean): the average distance between the first concave sub-region and the adjacent regions.

[0099] Distance standard deviation (Dσ): the standard deviation of the distance between the first concave sub-region and the adjacent regions.

[0100] Aspect ratio: the ratio of the width to the height of the first concave sub-region.

[0101] Tilt angle (θ): the angle between the longitudinal change and the lateral change of the pixel point in the first concave sub-region.

[0102] S4. Determine whether there is a second concave sub-region:

[0103] According to the spatial distribution characteristic values ​​of each first concave sub-region, it is judged whether there is a second concave sub-region in these regions. The second concave sub-region refers to a concave region that may be caused by factors such as vehicle overweight.

[0104] S41, obtaining the number of pixels in the second concave sub-region and determining the cause of the concave:

[0105] If there are second concave sub-areas, the number of pixels in these areas is obtained, and it is determined whether the concave is caused by the vehicle being overweight based on the number of pixels, which is achieved by comparing the number of pixels with a preset vehicle overweight concave characteristic value.

[0106] S5, removing the second concave sub-region to form a third concave sub-region:

[0107] The second concave sub-region in the first concave sub-region is removed, and the remaining part forms the third concave sub-region. These third concave sub-regions represent the real construction quality problem areas.

[0108] S6. Detect and analyze the distribution characteristics of the third concave sub-region:

[0109] The distribution characteristics of the third depression sub-area were tested and analyzed in detail, including parameters such as its area, depth, and shape complexity. Based on these analysis results, a highway construction quality inspection report was generated to provide a scientific basis for subsequent maintenance and repair.

[0110] The calculation of morphological feature values ​​includes the following steps:

[0111] Calculate the average gray value (G m ), the average gray value refers to the average gray value of all pixels in the concave area;

[0112] Calculate the edge density (E) of each initial concave sub-region d ), edge density refers to the density of edge pixels in the concave area;

[0113] Calculate the shape complexity (S) of each initial concave sub-region c ), shape complexity refers to the shape complexity of the concave area, which is determined by calculating the ratio of perimeter to area;

[0114] Calculate the area (A) of each initial concave sub-region, where the area refers to the total number of pixels in the concave region;

[0115] The morphological characteristic value is calculated by combining the above-mentioned characteristic values. The calculation formula of the morphological characteristic value is: F(t) = αG m +βE d +γS c +δA;

[0116] Among them, α, β, γ, and δ are weight coefficients, which are adjusted according to the specific application scenario;

[0117] The calculation formula of the average gray value is: Where N is the total number of pixels in the concave sub-region, G i is the gray value of the i-th pixel;

[0118] The edge density is calculated as follows: E is the total number of edge pixels in the concave sub-region, and A is the area of ​​the concave sub-region;

[0119] The calculation formula of the shape complexity is: Where P is the perimeter of the concave sub-region, and A is the area of ​​the concave sub-region;

[0120] The area of ​​the concave sub-region is calculated as follows: Where N is the total number of pixels in the concave sub-region;

[0121] The weight coefficient α is used to calculate the average gray value; the weight coefficient β is used to calculate the edge density; the weight coefficient γ is used to calculate the shape complexity; the weight coefficient δ is used to calculate the area;

[0122] The calculation method of α, β, γ, and δ is:

[0123]

[0124] Among them, μ G , μ E , μ S , μ A are the average gray value, edge density, shape complexity and area respectively, N is the total number of pixels in the concave sub-region, G i is the gray value of the i-th pixel, E j is the edge density of the jth sample; S k is the shape complexity of the kth sample; A l The lth step is to determine the spatial distribution characteristic value of each first concave sub-region according to the morphological characteristic value of each initial concave sub-region, each pixel point in each first concave sub-region and the position of each pixel point, including:

[0125] Get the adjacent areas of each first concave sub-area:

[0126] The adjacent region refers to a region that is in contact with or close to the first recessed sub-region.

[0127] According to the edge pixel positions of each first concave sub-region and its adjacent region, the distance feature of each first concave sub-region is determined:

[0128] Distance features include minimum distance, maximum distance, average distance, and standard deviation.

[0129] These distance features are determined by calculating the pixel distances between the concave sub-region and its adjacent regions.

[0130] According to the morphological feature values ​​of each initial concave sub-region, the morphological feature values ​​of each first concave sub-region are further refined:

[0131] By further analyzing the morphological feature values ​​of the initial concave sub-region, more accurate morphological features of the first concave sub-region are obtained.

[0132] According to the width and height of each first recessed sub-region and its adjacent region, the aspect ratio of each first recessed sub-region is determined:

[0133] The aspect ratio refers to the ratio of the width to the height of the concave sub-region.

[0134] Fitting processing is performed on all pixel points in each first concave sub-region and its adjacent region with the straight line equation to obtain the inclination angle of each first concave sub-region and its adjacent region:

[0135] The tilt angle is obtained by fitting the pixel points of the concave sub-region and its adjacent regions with the straight line equation.

[0136] According to the distance characteristics, morphological characteristic values, aspect ratios and inclination angles of each first concave sub-region, the spatial distribution characteristic values ​​of each first concave sub-region are determined:

[0137] By comprehensively analyzing these characteristic values, the spatial distribution characteristics of the first concave sub-region are determined. The area of ​​the sample;

[0138] The calculation of spatial distribution eigenvalues ​​includes the following steps:

[0139] B1. Calculate the minimum distance between the first concave sub-region and the adjacent region, and the calculation formula is: D min =min({d i}), where d i is the distance between the edge pixels of the first concave sub-region and the edge pixels of the adjacent region;

[0140] B2. Calculate the maximum distance between the first concave sub-region and the adjacent region, and the calculation formula is: D max =max({d i}), where d i is the distance between the edge pixels of the first concave sub-region and the edge pixels of the adjacent region;

[0141] B3. Calculate the average distance between the first concave sub-region and the adjacent region, and the calculation formula is:

[0142] where d i is the distance between the edge pixel of the first concave sub-region and the edge pixel of the adjacent region, and n is the total number of distances;

[0143] B4. Calculate the standard deviation of the distance between the first concave sub-region and the adjacent regions, and the calculation formula is: where d i is the distance between the edge pixel of the first concave sub-region and the edge pixel of the adjacent region, D mean is the average distance, n is the total number of distances;

[0144] B5. Calculate the aspect ratio of the first concave sub-region using the following calculation formula: Wherein, W is the width of the first recessed sub-region, and H is the height of the first recessed sub-region;

[0145] B6. Calculate the inclination angle of the first concave sub-region, and the calculation formula is: Wherein Δy is the longitudinal change of the pixel point in the first concave sub-area, and Δx is the lateral change;

[0146] B7. The morphological characteristic value F(t) is used as one of the input parameters to calculate the comprehensive spatial distribution characteristic value. The calculation formula is: S d =F(t)+D min +D max +D mean +D σ +R aspect +θ, where F(t) is the calculated morphological feature value;

[0147] Judging whether there is a second concave sub-region in each first concave sub-region according to the spatial distribution characteristic value of each first concave sub-region comprises the following steps:

[0148] The spatial distribution characteristic value S of the first concave sub-region d Compare with preset threshold:

[0149] By calculating the spatial distribution characteristic value S d The result is compared with a preset threshold to determine whether there is a potential second concave sub-region.

[0150] If S d exceeds a preset threshold, the first concave sub-region is marked as a potential second concave sub-region: If S d If the value is greater than a preset threshold, the first concave sub-region is marked as a potential second concave sub-region.

[0151] The first concave sub-region marked as a potential second concave sub-region is further analyzed to determine its specific location and range:

[0152] For the first depressed sub-region marked as having a potential second depressed sub-region, further detailed analysis is performed to determine its specific position and range.

[0153] If there is a second concave sub-region, the number of pixels in the second concave sub-region is obtained, and according to the number of pixels in the second concave sub-region, it is determined whether there is a concave caused by an overweight vehicle in the highway image, including the following steps:

[0154] Calculate the number of pixels in the second concave sub-area:

[0155] Count the number of all pixels in the second concave sub-region.

[0156] Compare the calculated results with the preset vehicle overweight dent characteristic value:

[0157] The number of pixels in the second concave sub-region is compared with a preset characteristic value.

[0158] If the number of pixels exceeds the preset value, it is judged that the area may be caused by overweight vehicles:

[0159] If the number of pixels exceeds a preset value, it is determined that the sunken area may be caused by an overweight vehicle.

[0160] Mark the judgment results on the highway image for subsequent processing and maintenance:

[0161] Mark the judgment results on the highway image to facilitate subsequent processing and maintenance;

[0162] The second recessed sub-region in the first recessed sub-region is removed to form a third recessed sub-region, comprising the following steps:

[0163] Mark and remove the second concave sub-region in the first concave sub-region:

[0164] All second concave sub-regions in the first concave sub-regions are marked and removed.

[0165] Recalculate the remaining concave area to form the third concave sub-area:

[0166] The remaining first concave sub-regions are recalculated to form third concave sub-regions.

[0167] Update the highway image to reflect the specific location and extent of the third depression sub-region:

[0168] Update the highway image to show the specific location and range of the third depression sub-area.

[0169] Analyzing the distribution characteristics of the third depressed sub-region includes the following steps:

[0170] Calculate the distribution characteristics of the third concave sub-region, including area, depth, and shape complexity:

[0171] The third concave sub-region is analyzed in detail and its area, depth and shape complexity are calculated.

[0172] Compare the calculated results with standard distribution characteristics to evaluate construction quality issues:

[0173] Compare the calculated results with preset standard distribution characteristics to evaluate construction quality issues.

[0174] Generates a detailed inspection report with specific information and maintenance recommendations for the third depression sub-area:

[0175] Generates inspection reports with detailed information and maintenance recommendations.

[0176] Store the test report in the database for subsequent query and analysis:

[0177] The test report is stored in the database for subsequent query and analysis.

[0178] Example 1: Used for highway construction quality inspection

[0179] Background

[0180] During highway construction, the quality of construction directly affects the performance and service life of the highway. Especially on highways, road surface quality problems may lead to serious traffic accidents. Therefore, it is very important to detect construction quality efficiently and accurately.

[0181] This embodiment aims to detect quality problems generated during highway construction by using image processing technology and computer vision algorithms, especially to identify sunken areas caused by poor construction quality.

[0182] Innovation: The method of comprehensive spatial distribution eigenvalues ​​is adopted to combine traditional image processing with modern computer vision algorithms to improve detection accuracy and efficiency.

[0183] Application scenarios: This method is suitable for real-time quality inspection at highway construction sites, especially for comprehensive quality assessment after construction is completed.

[0184] Implementation steps

[0185] Image preprocessing: Obtain highway images at the construction site and perform denoising, contrast enhancement, and distortion correction.

[0186] Determination of initial concave sub-region: The initial concave sub-region is determined by edge detection and gray value change.

[0187] Morphological feature value calculation: The morphological feature value of the initial concave sub-region is calculated based on the gray value, edge density, shape complexity and area.

[0188] Calculation of spatial distribution characteristic values: Further calculation of spatial distribution characteristic values, including minimum distance, maximum distance, average distance, standard deviation, aspect ratio and tilt angle.

[0189] Determine the second sunken sub-region: compare the spatial distribution characteristic value with a preset threshold to determine whether there is a sunken region caused by the overweight vehicle.

[0190] Eliminating the second concave sub-region: eliminating the second concave sub-region in the first concave sub-region to form a third concave sub-region.

[0191] Third depression sub-region analysis: Calculate the distribution characteristics of the third depression sub-region and generate a detailed detection report.

[0192] Application effect: This method can quickly and accurately identify construction quality problems in highway construction quality inspection, and provide a detailed inspection report to provide a scientific basis for subsequent maintenance.

[0193] Compared with traditional methods, the present invention greatly improves detection accuracy, reduces the misjudgment rate, and has the characteristics of strong real-time performance.

[0194] Comparative experiment:

[0195] project Traditional methods Method of the present invention Detection accuracy 80％ 95％ False positive rate 20％ 5％ Detection speed slow quick Real-time Difference powerful

[0196] Example 2: Used for municipal road construction quality inspection

[0197] Background

[0198] The quality of municipal roads directly affects the smoothness of urban traffic and the safety of citizens' travel. Poor construction quality will lead to problems such as road dents and cracks, affecting the overall image of the city and traffic efficiency.

[0199] Technical Purpose

[0200] This embodiment aims to detect quality problems in municipal road construction through efficient image processing and analysis methods, especially to identify depressions and cracks caused by construction quality problems.

[0201] Innovation: It combines a variety of image processing technologies and spatial distribution eigenvalue analysis to improve the detection accuracy of dents and cracks.

[0202] Application scenario: This method is suitable for comprehensive quality inspection after municipal road construction, especially for quality acceptance before the road is put into use.

[0203] Implementation steps

[0204] Image preprocessing: Obtain images of municipal roads and perform denoising, contrast enhancement, and distortion correction.

[0205] Determination of initial concave sub-region: The initial concave sub-region is determined by edge detection and gray value change.

[0206] Morphological feature value calculation: The morphological feature value of the initial concave sub-region is calculated based on the gray value, edge density, shape complexity and area.

[0207] Calculation of spatial distribution characteristic values: Further calculation of spatial distribution characteristic values, including minimum distance, maximum distance, average distance, standard deviation, aspect ratio and tilt angle.

[0208] Determine the second sunken sub-region: compare the spatial distribution characteristic value with a preset threshold to determine whether there is a sunken region caused by the overweight vehicle.

[0209] Eliminating the second concave sub-region: eliminating the second concave sub-region in the first concave sub-region to form a third concave sub-region.

[0210] Third depression sub-region analysis: Calculate the distribution characteristics of the third depression sub-region and generate a detailed detection report.

[0211] This method can quickly and accurately identify quality problems in municipal road construction quality inspection and provide a detailed inspection report, providing a scientific basis for subsequent maintenance and acceptance.

[0212] Compared with traditional methods, the present invention greatly improves detection accuracy, reduces the misjudgment rate, and can promptly detect and resolve quality problems before the road is put into use.

[0213] Comparative experiment:

[0214] project Traditional methods Method of the present invention Detection accuracy 75％ 92％ False positive rate 25％ 8％ Detection speed slow quick Real-time Difference powerful

[0215] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A highway construction quality detection method, characterized in that: The following steps are involved: S1, obtaining a highway image that has undergone image preprocessing, including the steps of denoising, contrast enhancement, and distortion correction, and determining each initial concave sub-region in the highway image according to the highway image; S2, determining the morphological feature value of each initial concave sub-region according to the gray value of each pixel point in each initial concave sub-region and the number of pixels points in each initial concave sub-region; S3, judging whether there is a first concave sub-region in each of the initial concave sub-regions according to the morphological feature values ​​of each of the initial concave sub-regions; If so, determining the spatial distribution characteristic value of each first concave sub-region according to the morphological characteristic value of each initial concave sub-region, each pixel point in each first concave sub-region and the position of each pixel point; S4, judging whether there is a second concave sub-region in each of the first concave sub-regions according to the spatial distribution characteristic values ​​of each of the first concave sub-regions; If yes, then obtaining the number of pixels in the second concave sub-region, and judging whether there is a concave caused by the overweight vehicle in the highway image according to the number of pixels in the second concave sub-region; S5, removing the second concave sub-region from the first concave sub-region to form a third concave sub-region; S6. Detect and analyze the distribution characteristics of the third concave sub-region.

2. A highway construction quality detection method according to claim 1, characterized in that: The calculation of the morphological feature value comprises the following steps: Calculate the average gray value (G m ); Calculate the edge density (E) of each initial concave sub-region d ); Calculate the shape complexity (S) of each initial concave sub-region c ); Calculate the area (A) of each initial concave sub-region; The calculation formula of the morphological characteristic value is: F(t) = αG m +βE d +γS c +δA; Among them, α, β, γ, and δ are weight coefficients, which are adjusted according to the specific application scenario.

3. A highway construction quality detection method according to claim 2, characterized in that: The calculation formula of the average gray value is: Where N is the total number of pixels in the concave sub-region, G i is the gray value of the i-th pixel; The edge density is calculated as follows: E is the total number of edge pixels in the concave sub-region, and A is the area of ​​the concave sub-region; The calculation formula of the shape complexity is: Where P is the perimeter of the concave sub-region, and A is the area of ​​the concave sub-region; The area of ​​the concave sub-region is calculated as follows: Wherein, N is the total number of pixels in the concave sub-region.

4. A highway construction quality detection method according to claim 3, characterized in that: The weight coefficients α, β, γ, and δ are calculated as follows: Among them, μ G , μ E , μ S , μ A are the average gray value, edge density, shape complexity and area respectively, and N, M, P, Q are the total number of samples of the corresponding eigenvalues ​​respectively.

5. A highway construction quality detection method according to claim 4, characterized in that: Determining the spatial distribution characteristic value of each first concave sub-region according to the morphological characteristic value of each initial concave sub-region, each pixel point in each first concave sub-region and the position of each pixel point includes: Acquire adjacent regions of each first concave sub-region; Determine the distance characteristics of each first concave sub-region according to the edge pixel positions of each first concave sub-region and its adjacent region, including the minimum distance, the maximum distance, the average distance and the standard deviation; Further refining the morphological feature values ​​of each first concave sub-region according to the morphological feature values ​​of each initial concave sub-region; Determining the aspect ratio of each first recessed sub-region according to the width and height of each first recessed sub-region and its adjacent region; Fitting all pixel points in each first concave sub-region and its adjacent region with a straight line equation to obtain the inclination angle of each first concave sub-region and its adjacent region; The spatial distribution characteristic value of each first recessed sub-region is determined according to the distance characteristic, the morphological characteristic value, the aspect ratio and the inclination angle of each first recessed sub-region.

6. A highway construction quality detection method according to claim 5, characterized in that: The calculation of the spatial distribution characteristic value comprises the following steps: B1. Calculate the minimum distance between the first concave sub-region and the adjacent region, and the calculation formula is: D min =min({d i }), where d i is the distance between the edge pixels of the first concave sub-region and the edge pixels of the adjacent region; B2. Calculate the maximum distance between the first concave sub-region and the adjacent region, and the calculation formula is: D max =max({d i }), where d i is the distance between the edge pixels of the first concave sub-region and the edge pixels of the adjacent region; B3. Calculate the average distance between the first concave sub-region and the adjacent region, and the calculation formula is: where d i is the distance between the edge pixel of the first concave sub-region and the edge pixel of the adjacent region, and n is the total number of distances; B4. Calculate the standard deviation of the distance between the first concave sub-region and the adjacent regions, and the calculation formula is: where d i is the distance between the edge pixel of the first concave sub-region and the edge pixel of the adjacent region, D mean is the average distance, n is the total number of distances; B5. Calculate the aspect ratio of the first concave sub-region using the following calculation formula: Wherein, W is the width of the first recessed sub-region, and H is the height of the first recessed sub-region; B6. Calculate the inclination angle of the first concave sub-region, and the calculation formula is: Wherein Δy is the longitudinal change of the pixel point in the first concave sub-area, and Δx is the lateral change; B7. The morphological characteristic value F(t) is used as one of the input parameters to calculate the comprehensive spatial distribution characteristic value. The calculation formula is: S d =F(t)+D min +D max +D mean +D σ +R aspect +θ, where F(t) is the calculated morphological feature value.

7. A highway construction quality detection method according to claim 1, characterized in that: Judging whether there is a second concave sub-region in each first concave sub-region according to the spatial distribution characteristic value of each first concave sub-region comprises the following steps: The spatial distribution characteristic value S of the first concave sub-region d Compare with a preset threshold; If S d If the threshold is exceeded, the first concave sub-region is marked as a potential second concave sub-region; The first depressed sub-region marked as a potential second depressed sub-region is further analyzed to determine its specific position and range.

8. A highway construction quality detection method according to claim 1, characterized in that: If there is a second concave sub-region, the number of pixels in the second concave sub-region is obtained, and according to the number of pixels in the second concave sub-region, it is determined whether there is a concave caused by an overweight vehicle in the highway image, including the following steps: Calculating the number of pixels in the second concave sub-region; Compare the calculated result with a preset vehicle overweight depression characteristic value; If the number of pixels exceeds the preset value, it is determined that the area may be caused by an overweight vehicle; The judgment results are marked on the highway image for subsequent processing and maintenance.

9. A highway construction quality detection method according to claim 1, characterized in that: The second recessed sub-region in the first recessed sub-region is removed to form a third recessed sub-region, comprising the following steps: Mark and remove the second concave sub-region in the first concave sub-region; Recalculating the remaining concave area to form a third concave sub-area; The highway image is updated to reflect the specific location and range of the third concave sub-region.

10. A highway construction quality detection method according to claim 1, characterized in that: Analyzing the distribution characteristics of the third depressed sub-region includes the following steps: Calculating the distribution characteristics of the third concave sub-region, including area, depth and shape complexity; Compare the calculated results with the standard distribution characteristics to evaluate construction quality issues; Generate a detailed inspection report with specific information and maintenance recommendations for the third depression sub-area; The test report is stored in the database for subsequent query and analysis.

Images