A municipal road construction safety monitoring method

Abstract

The present invention relates to the field of image processing technology, and in particular to a municipal road construction safety monitoring method, the method comprising: obtaining a road image, preprocessing the road image, obtaining a road surface settlement image; obtaining a discontinuity area in the road surface settlement image; obtaining a waterlogged discontinuity area in the discontinuity area; repairing cracks in the waterlogged discontinuity area to obtain complete cracks corresponding to the waterlogged discontinuity area; and classifying the road surface quality of the road according to the repair result. The present invention can improve the accuracy of road quality detection.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a municipal road construction safety monitoring method. Background Art

[0002] The most commonly used road construction safety monitoring in life is to install cameras to achieve real-time image monitoring, helping managers to promptly discover and deal with safety hazards. This method can improve the safety management efficiency of the construction site, and can also improve road quality by timely discovering construction quality problems through real-time monitoring and data analysis.

[0003] However, when the existing edge detection technology is used to locate the interrupted area of ​​road subsidence cracks blocked by water accumulation, the blurred monitoring pixels will cause the crack area in the edge detection image to be interrupted. Such interrupted areas will interfere with the judgment of the interrupted areas caused by water accumulation on the road, thereby affecting the detection effect. Summary of the invention

[0004] In order to solve the technical problem that road surface water affects the road quality detection effect, the purpose of the present invention is to provide a municipal road construction safety monitoring method, and the technical solution adopted is as follows:

[0005] In a first aspect of the present disclosure, a method for monitoring the safety of municipal road construction is provided, the method comprising:

[0006] Acquire an image of a road, pre-process the image of the road, and acquire a road surface settlement image;

[0007] Acquire a discontinuity region in the road surface settlement image;

[0008] Acquire a waterlogging interruption area in the interruption area;

[0009] Repairing the cracks in the water accumulation interruption area to obtain complete cracks corresponding to the water accumulation interruption area;

[0010] According to the repair results, the road surface quality of the road is classified.

[0011] In one embodiment, the preprocessing of the road image to obtain the road subsidence image includes:

[0012] Acquire a subsidence area image in the image of the road;

[0013] grayscale the subsidence area image to obtain a subsidence area grayscale image;

[0014] Performing edge detection on the grayscale image of the sedimentation area to obtain an edge detection image;

[0015] The road surface subsidence image in the subsidence area image is acquired according to the edge detection image.

[0016] In one embodiment, the obtaining of the interrupted area in the road surface settlement image includes:

[0017] Extracting crack pixel points corresponding to the crack area in the road surface settlement image;

[0018] Performing region growth on the crack pixel points to obtain multiple crack growth regions;

[0019] The following steps are performed for each of the crack growth regions:

[0020] In the other crack growth regions, according to the last pixel point of the crack interruption end in the current crack growth region and the crack end pixel points in the other crack growth regions, a candidate crack growth region belonging to the same crack as the current crack growth region is obtained;

[0021] In the candidate crack growth region, according to the crack width of the current crack growth region and the crack width of the candidate crack growth region, a target crack growth region belonging to the same crack as the current crack growth region is obtained;

[0022] The portion between the current crack growth region and the target crack growth region is the interruption region.

[0023] In one embodiment, in the other crack growth regions, obtaining a candidate crack growth region belonging to the same crack as the current crack growth region according to the last pixel point of the crack interruption end in the current crack growth region and the crack end pixel points of the other crack growth regions includes:

[0024] Obtaining the last pixel point of the crack break end in the current crack growth area;

[0025] Get the previous pixel point adjacent to the last pixel point;

[0026] Obtaining the angle between a pixel vector and each candidate vector, wherein the pixel vector is a vector formed by the previous pixel and the last pixel, and each candidate vector is a vector formed by the last pixel and each other pixel of the crack end of the crack growth area;

[0027] Obtaining the modulus length of each of the candidate vectors;

[0028] In each of the included angles and the corresponding modulus lengths, a candidate crack end pixel point that meets a first preset threshold is obtained;

[0029] Among them, the crack growth region corresponding to the pixel point at the end of the candidate crack is the candidate crack growth region.

[0030] In one embodiment, in the candidate crack growth region, obtaining a target crack growth region belonging to the same crack as the current crack growth region according to the crack width of the current crack growth region and the crack width of the candidate crack growth region includes:

[0031] Calculating a first average value of crack widths within a preset length from the last pixel point in the current crack growth region;

[0032] A second average value of crack width within the preset length from the pixel point at the end of the candidate crack in each of the candidate crack growth regions is calculated.

[0033] According to the first average value and each of the second average values, a target crack end pixel point that meets a second preset threshold is obtained from the candidate crack end pixel points, and the candidate crack growth region where the target crack end pixel point is located is the target crack growth region.

[0034] In one embodiment, obtaining the waterlogging interruption area in the interruption area includes:

[0035] Establishing a first minimum bounding rectangle for each of the interruption areas, and obtaining the area of ​​each of the first minimum bounding rectangles;

[0036] Obtain the sum of the areas of all the first minimum circumscribed rectangles;

[0037] Calculate the average grayscale value of all pixels in each of the first minimum circumscribed rectangles;

[0038] Obtaining the sum of average grayscale values ​​of all the first minimum bounding rectangles;

[0039] The water accumulation interruption area in the interruption area is obtained according to the area of ​​each of the first minimum circumscribed rectangles, the sum of the areas, the average grayscale value of each of the first minimum circumscribed rectangles, and the sum of the average grayscale values.

[0040] In one embodiment, the step of obtaining the waterlogged interruption area in the interruption area according to the area of ​​each of the first minimum circumscribed rectangles, the sum of the areas, the average grayscale value of each of the first minimum circumscribed rectangles, and the sum of the average grayscale values ​​includes:

[0041] For each of the first minimum bounding rectangles, perform the following steps:

[0042] According to the area of ​​the current first minimum circumscribed rectangle and the sum of the areas, obtaining a first possible value of the interruption area corresponding to the current first minimum circumscribed rectangle becoming the waterlogged interruption area;

[0043] According to the average grayscale value of the current first minimum circumscribed rectangle and the sum of the average grayscale values, obtaining a second possible value of the interruption area corresponding to the current first minimum circumscribed rectangle becoming the waterlogged interruption area;

[0044] Obtaining a final possible value for the interruption area corresponding to the current first minimum circumscribed rectangle to become the waterlogged interruption area according to the first possible value and the second possible value;

[0045] When the final possible value is greater than a third preset threshold, it is determined that the interruption area corresponding to the current first minimum circumscribed rectangle is the water accumulation interruption area.

[0046] In one embodiment, the repairing of the cracks in the water accumulation interruption area to obtain a complete crack corresponding to the water accumulation interruption area includes:

[0047] For each of the water accumulation interruption areas, obtaining the last pixel point corresponding to the water accumulation interruption area and the target crack end pixel point;

[0048] The last pixel point in each of the water accumulation interruption areas is connected to the pixel point at the end of the target crack to obtain a complete crack corresponding to each of the water accumulation interruption areas.

[0049] In one embodiment, the classifying the road surface quality of the road according to the repair result includes:

[0050] Obtaining the entire crack region according to the complete cracks corresponding to all the water interruption regions and all the crack growth regions;

[0051] Establishing a second minimum circumscribed rectangle for the entire crack region;

[0052] The road surface quality of the road is classified according to the pixel ratio of the crack area in the second minimum circumscribed rectangle and the area of ​​the second minimum circumscribed rectangle.

[0053] In one embodiment, classifying the road surface quality of the road according to the pixel point ratio of the crack area in the second minimum circumscribed rectangle and the area of ​​the second minimum circumscribed rectangle includes:

[0054] Obtaining the product of the pixel ratio of the crack area in the second minimum circumscribed rectangle and the area of ​​the second minimum circumscribed rectangle;

[0055] When it is detected that the product is less than a first threshold, determining that the road surface damage degree of the road is that the road surface begins to settle;

[0056] When it is detected that the product is less than or equal to the second threshold and greater than or equal to the first threshold, determining that the road surface damage degree of the road is that settlement has occurred;

[0057] When it is detected that the product is greater than the second threshold, it is determined that the road surface damage degree of the road is about to collapse.

[0058] The present invention has the following beneficial effects:

[0059] The present disclosure studies the problem that cracks caused by road surface settlement due to water accumulation are interrupted in edge detection results. First, the interrupted area of ​​the road surface in the captured road image is located, and then the water accumulation interrupted area is further located in the located interrupted area, and then the cracks in the water accumulation interrupted area are repaired to obtain the complete cracks corresponding to the water accumulation interrupted area, and finally the road surface quality of the road is classified according to the repair results. Specifically, the settlement area where water accumulates on the road will form a reflective area, then this reflective area appears as a higher gray value area on the computer vision edge detection image, and this area interrupts the cracks in the settlement area in the edge detection. The present disclosure mainly locates the interrupted area according to its characteristics, and repairs the interrupted area according to the extension direction of the pixel points after determining the position. The present disclosure solves the impact of road water accumulation on road settlement characteristics that cannot be avoided based on edge detection technology, and improves the accuracy of road quality detection to a certain extent. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0061] Figure 1 A process of a municipal road construction safety monitoring method provided by an embodiment of the present invention Figure 1 ;

[0062] Figure 2 A process of a municipal road construction safety monitoring method provided by an embodiment of the present invention Figure 2 ;

[0063] Figure 3 An image of a road provided by one embodiment of the present invention;

[0064] Figure 4 An embodiment of the present invention provides Figure 3 An image of a subsidence area in an image of a road;

[0065] Figure 5 An embodiment of the present invention provides Figure 4 A grayscale image of the subsidence area of ​​the subsidence area image in FIG.

[0066] Figure 6 An embodiment of the present invention provides Figure 5 The edge detection image corresponding to the grayscale image of the subsidence area in ;

[0067] Figure 7 A process of a municipal road construction safety monitoring method provided by an embodiment of the present invention Figure 3 ;

[0068] Figure 8 A process of a municipal road construction safety monitoring method provided by an embodiment of the present invention Figure 4 ;

[0069] Fig. 9 The present invention is a block diagram of a municipal road construction safety monitoring device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0070] 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 a municipal road construction safety monitoring method proposed by the present invention, its specific implementation method, structure, features and effects, in conjunction 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.

[0071] 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.

[0072] The specific scheme of a municipal road construction safety monitoring method provided by the present invention is described in detail below with reference to the accompanying drawings.

[0073] See also Figure 1 , which shows a flow chart of a municipal road construction safety monitoring method provided by an embodiment of the present invention, comprising the following steps S101-S105:

[0074] S101, obtaining a road image, preprocessing the road image, and obtaining a road surface settlement image.

[0075] In one embodiment, Figure 2 As shown, preprocessing the road image to obtain the road subsidence image includes the following sub-steps S1011-S1014:

[0076] S1011. Acquire a subsidence area image in an image of the road.

[0077] The road can be photographed by surveillance cameras above the road surface, and the image information is transmitted to the computer, which generates a digital picture through relevant technology. The image can accurately reflect the relevant conditions of the road surface by using the shape and color of objects, such as Figure 3 . Figure 3 There are cracks in some areas of the road that are about to sink. In order to observe the corresponding crack areas in more detail, the image is enlarged here to obtain Figure 4 Image of the middle subsidence area, where Figure 3 and Figure 4 Both can be color images.

[0078] S1012, grayscale processing is performed on the subsidence area image to obtain a subsidence area grayscale map.

[0079] In order to facilitate the computer to identify and analyze the road defect area in the image, Figure 4 The sedimentation area image in the grayscale processing is obtained Figure 5 Grayscale image of the subsidence area in .

[0080] S1013, performing edge detection on the grayscale image of the subsidence area to obtain an edge detection image.

[0081] right Figure 5 The grayscale image of the settlement area in the image is used for edge detection to identify the relevant features of the road settlement, and the following are obtained: Figure 6 The edge detection diagram shown.

[0082] S1014. Obtain a road surface subsidence image in the subsidence area image according to the edge detection image.

[0083] In order to achieve the division between grass and road surface, Figure 3 It can be seen that the grass area on one side of the curb has more entities and more complex shapes and contours, while the road surface area on the other side of the curb has fewer entities because the road surface is relatively flat.

[0084] Based on the large differences between the two sides of the curb, in the edge detection image, the edge detection contour on the grass side is more complex, and more detection targets are reflected, while the result of the road surface area is just the opposite. Here, the road surface image can be determined by calculating the number of pixels in the two areas and comparing the number of pixels. The side with more pixels is the grass, and the side with less pixels is the road surface image. The road surface image here represents the above-mentioned road subsidence image.

[0085] S102: Acquire the interruption area in the road subsidence image.

[0086] If the force on a certain part of the road surface is unbalanced due to the loss or corrosion of some materials at the bottom, then the force at that point is not enough to support the weight of the entire road surface, which will cause tension on the road surface and form initial cracks. The settled road surface will form low-lying areas, making it easy for rainwater or other water sources to gather and form water accumulation. These accumulated water will form reflective areas under the surveillance camera angle, and these reflective areas will cover the crack areas caused by the road surface settlement.

[0087] First, the interruption area will be preliminarily determined. The corresponding road crack in the edge detection result is interrupted, and the interrupted area is large, the crack is long, and is distributed in sheets. There is a certain amount of interruption in the crack area. The main content of this step is to use the interruption area features to find the corresponding interruption area, which specifically includes the following sub-steps A1-A4:

[0088] A1. Extract crack pixel points corresponding to the crack area in the road settlement image.

[0089] A2. Perform regional growth on crack pixels to obtain multiple crack growth regions.

[0090] After finding the crack pixel points, regional growth is performed on these crack pixel points, and a crack growth region composed of multiple grown strip-shaped pixel points representing cracks is obtained here. Based on the grayscale value difference between the crack pixel point and the crack pixel points in the neighborhood, regional growth is performed on the crack pixel point to obtain multiple crack growth regions. Among them, the specific process of regional growth of the crack pixel point is to use any crack pixel point as the initial seed point, and perform regional growth when the grayscale value difference between the initial seed point and the crack pixel points in the 8 neighborhoods is less than the preset grayscale threshold, that is, the initial seed point that meets the threshold requirement and the crack pixel points in the 8 neighborhoods are classified as the same crack growth region. It should be noted that the preset grayscale threshold is used to measure the maximum grayscale difference between adjacent pixels during the regional growth process, and the regional growth algorithm is a well-known technology and will not be introduced in detail here.

[0091] For each crack growth region, perform the following steps A3-A4:

[0092] A3. In other crack growth regions, based on the last pixel point of the crack break in the current crack growth region and the crack end pixel points of other crack growth regions, obtain candidate crack growth regions that belong to the same crack as the current crack growth region.

[0093] In one embodiment, A3 includes the following sub-steps A31-A35:

[0094] A31. Obtain the last pixel point of the crack break end in the current crack growth area.

[0095] When the crack in the current crack growth region is a crack from top to bottom, at this time, from top to bottom, locate the endpoint of the tail of the crack region, where the endpoint a of the tail of the crack region represents the last pixel point a of the crack interruption end in the above embodiment. When the crack in the current crack growth region is a crack from left to right, at this time, from left to right, locate the rightmost endpoint of the crack region, where the rightmost endpoint a of the crack region represents the last pixel point a of the crack interruption end in the above embodiment.

[0096] A32. Get the previous pixel point adjacent to the last pixel point.

[0097] A33. Obtain the angle between the pixel vector and each candidate vector. The pixel vector is a vector formed by the previous pixel and the last pixel. Each candidate vector is a vector formed by the last pixel and the crack end pixel points of other crack growth areas.

[0098] Find the last pixel point a at the end of the crack and the adjacent previous pixel point b along the crack direction, and construct the pixel point vector .

[0099] In the cracks of other crack growth regions, for each crack end pixel c in the crack growth region, a candidate vector is constructed , and calculate the pixel vector With each candidate vector The angle β is Represents the candidate vector Vector with pixels The possibility of a unified crack, and the smaller the angle β, the The larger the value, the greater the possibility that the two vectors are in the same direction, that is, the greater the possibility that the crack at the pixel point c is the same crack as the crack at the pixel point a.

[0100] A34. Obtain the modulus length of each candidate vector.

[0101] A35. In each angle and corresponding modulus, a candidate crack end pixel point that meets a first preset threshold is obtained; wherein the crack growth region corresponding to the candidate crack end pixel point is the candidate crack growth region.

[0102] Here, in order to avoid pixel point a matching the crack end pixel of the next crack in the same direction, further screening is achieved by setting the distance. Specifically: each candidate vector can be calculated The modulus length d is used to judge the possibility of each candidate vector. The smaller the modulus length, the more likely the candidate vector is. The greater the crack end, the greater the possibility that the crack at pixel point c and the crack at pixel point a are the same crack.

[0103] Combined with the above angle, the candidate vector The possibility that the crack at pixel c is the same crack as the crack at pixel a is expressed as the following formula:

[0104]

[0105] In this formula, Represents the candidate vector The probability that the crack at the crack end pixel c and the crack at the pixel a are the same crack; Represents pixel vector With candidate vector The cosine value of the angle between the last pixel point a at the crack end and the adjacent previous pixel point b along the crack direction constitute the pixel vector The last pixel point a at the crack break end and the pixel point c at the crack end of other crack growth areas are constructed as the candidate vector ; Represents the candidate vector The linear normalization result of the modulus is between [0,1]. If the normalized result is closer to 0, it means that the value of the modulus before normalization is smaller.

[0106] In this formula, The larger the value, the more likely the candidate vector is to be The greater the probability that the crack at the crack end pixel c and the crack at the pixel a are the same crack, that is, the candidate vector The greater the possibility that the crack end pixel c is the endpoint of the interruption area where the pixel a is located.

[0107] The first preset threshold is used to screen the possibility that the cracks at two different crack end pixel points are the same crack. Through the value range of the possibility that the crack at the crack end pixel point c and the crack at the pixel point a are the same crack, the value range of the first preset threshold can be determined to be (0,1). When the value of the first preset threshold is closer to 0, it means that the standard for screening the crack end pixel points using the first preset threshold is looser. When the value of the first preset threshold is closer to 1, it means that the standard for screening the crack end pixel points using the first preset threshold is stricter. A specific example is given in this embodiment, that is, this embodiment sets a stricter screening standard to determine the candidate crack end pixel points.

[0108] For example, the first preset threshold is set to 0.8, then all crack end pixel points c corresponding to when f is greater than 0.8 are candidate crack end pixel points, and the crack growth areas corresponding to the candidate crack end pixel points are candidate crack growth areas.

[0109] A4. In the candidate crack growth region, based on the crack width of the current crack growth region and the crack width of the candidate crack growth region, obtain a target crack growth region that belongs to the same crack as the current crack growth region; wherein the portion between the current crack growth region and the target crack growth region is the interruption region.

[0110] The cracks in the road subsidence area are covered by water on the road surface, and show a tendency to be interrupted in the image. When locating the interrupted area, there are some shorter cracks in other areas of the road surface. The tails of these cracks and the starting ends of other cracks constitute an area approximate to the interrupted area. These areas interfere with the positioning of the interrupted area with water in the present disclosure. The widths of the crack areas at both ends of the interrupted area with water have certain approximate characteristics and certain continuous characteristics. This step further judges the interrupted area with water based on these characteristics.

[0111] In one embodiment, step A4 includes the following sub-steps A41-A43:

[0112] A41. Calculate the first average value of the crack width within a preset length from the last pixel point under the straight line where the crack direction is located in the current crack growth area.

[0113] A42. Calculate the second average value of the crack width within a preset length from the pixel point at the end of the candidate crack under the straight line where the crack direction is located in each candidate crack growth region.

[0114] A43. According to the first average value and each second average value, a target crack end pixel point that meets a second preset threshold is obtained from the candidate crack end pixel points, and the candidate crack growth region where the target crack end pixel point is located is the target crack growth region.

[0115] After all candidate crack growth regions of the current crack growth region are obtained, if there is a water interruption region between the last pixel point of the current crack growth region and the candidate crack end pixel point, then the crack growth regions corresponding to these two pixels will have the characteristic that the closer they are to the water interruption region, the closer their widths are. First, the first average value L of the crack width within x lengths of the last pixel point a close to the current crack growth region is calculated; secondly, the second average value H of the crack width within x lengths of the candidate crack end pixel point c of the candidate crack growth region where the candidate crack end pixel point c is located is calculated, where the x lengths represent the above-mentioned preset lengths.

[0116] The preset length represents the value of the number of pixels close to the candidate crack end pixel point c in the candidate crack growth area where the candidate crack end pixel point c is located, and also represents the value of the number of pixels close to the last pixel point a in the current crack growth area, reflecting the length of the window containing the target.

[0117] Calculate the first average value of the crack width within a preset length from the last pixel point under the straight line where the crack direction is located in the current crack growth area; calculate the second average value of the crack width within the preset length from the candidate crack end pixel point under the straight line where the crack direction is located in each candidate crack growth area. The straight line where the crack direction is located refers to the line connecting the crack end pixel points in the crack growth area.

[0118] It can be understood that the difference between the average values ​​of the crack widths calculated by selecting pixels of a preset length is to analyze the area where the tail of the crack and the starting end of other cracks constitute an approximate interruption area, and the width has certain approximate characteristics. Therefore, it can be seen that the crack width of the preset length closest to the target is selected for feature analysis in the crack direction, which can avoid the large error of using a single crack width for difference analysis. It should be understood by those skilled in the art that, similar to the concept of neighborhood range, the value setting of the preset length can be understood as the number of pixels within the neighborhood range of the candidate crack end pixel c or the last pixel a in the current crack growth area. The implementer can set it according to the specific implementation scenario, such as 4 or 8, etc.

[0119] Among them, the first average value L of the crack width within x lengths of the last pixel point a close to the current crack growth area can be obtained by obtaining the number of pixel points in each column within x lengths in the current crack growth area, then summing the number of pixel points in each column, and then dividing it by the number of columns of pixel points within x lengths. The method for obtaining the second average value H is similar.

[0120] Then, the width values ​​of the two cracks corresponding to the two pixels are used to indicate the possibility that the two pixels are water interruption areas. The formula is as follows:

[0121] ;

[0122] In the formula, L represents the first average crack width within x lengths of the last pixel point a close to the current crack growth region, and H represents the second average crack width within x lengths of the candidate crack growth region close to the candidate crack end pixel point c where the candidate crack end pixel point c is located. Represents the normalization function. represents the difference between these two means, The greater the possibility that the last pixel point a in the current crack growth area and the candidate crack end pixel point c in the candidate crack growth area are the same crack.

[0123] For example, the second preset threshold can be set to 0.8. When j is greater than 0.8, the corresponding candidate crack end pixel point c represents the above-mentioned target crack end pixel point c. At this time, it can be determined that the last pixel point a in the current crack growth area and the target crack end pixel point c in the target crack growth area are pixels in the same crack, and the part between the current crack growth area and the target crack growth area represents the interruption area.

[0124] Based on similar reasons to the first preset threshold setting method, the second preset threshold is used to screen whether the last pixel point a in the current crack growth area and the target crack end pixel point c in the target crack growth area are pixels in the same crack, and then the value range of the second preset threshold can be determined to be (0,1). When the value of the second preset threshold is closer to 0, it means that the standard for screening whether the pixel points belong to the same crack using the second preset threshold is looser, and when the value of the second preset threshold is closer to 1, it means that the standard for screening whether the pixel points belong to the same crack using the second preset threshold is stricter. A specific example is given in this embodiment, that is, this embodiment requires a stricter screening standard to determine the candidate crack end pixel points.

[0125] It should be further explained that the values ​​set for the first preset threshold and the second preset threshold in this embodiment are the same, that is, the screening criteria concerned in this embodiment are consistent, and this embodiment provides an exemplary value with relatively good effect, which is set to 0.8 to avoid setting the screening criteria too loose or too strict.

[0126] S103, obtaining a waterlogging interruption area in the interruption area.

[0127] When judging the interruption area in the previous step, the pixel blur during the monitoring process will also cause the crack in the final edge detection image to be interrupted. Therefore, judging the interruption area based solely on the crack interruption feature in the previous step cannot accurately locate the interruption area of ​​water accumulation.

[0128] The area of ​​the waterlogged interruption area is larger than other interruption areas, and the area produces reflections due to the waterlogging, and the grayscale value of the reflective area is larger than other areas. Based on these two features, the interruption area determined by the above process can be confirmed as a waterlogged interruption area. Figure 7 As shown, the specific steps are as follows S1031-S1035:

[0129] S1031: Create a first minimum bounding rectangle for each interruption area, and obtain the area of ​​each first minimum bounding rectangle.

[0130] Establish the first minimum enclosing rectangle for each interruption area, and obtain the area of ​​each first minimum enclosing rectangle as .

[0131] S1032. Obtain the sum of the areas of all first minimum circumscribed rectangles.

[0132] S1033. Calculate the average grayscale value of all pixels in each first minimum circumscribed rectangle.

[0133] For each first minimum enclosing rectangle, obtain the number of pixels in the current first minimum enclosing rectangle and the grayscale value of each pixel in the current first minimum enclosing rectangle, sum the grayscale values ​​of each pixel in the current first minimum enclosing rectangle, and divide the result by the number of pixels in the current first minimum enclosing rectangle to obtain the average grayscale value of all pixels in the current first minimum enclosing rectangle. .

[0134] S1034. Obtain the sum of the average grayscale values ​​of all first minimum circumscribed rectangles.

[0135] S1035. Obtain a waterlogged interruption area in the interruption area according to the area of ​​each first minimum circumscribed rectangle, the sum of the areas, the average grayscale value of each first minimum circumscribed rectangle, and the sum of the average grayscale values.

[0136] Specifically, the following steps S10351-S10354 are performed for each first minimum circumscribed rectangle:

[0137] S10351. Obtain, based on the area of ​​the current first minimum circumscribed rectangle and the sum of the areas, a first possible value of the interruption area corresponding to the current first minimum circumscribed rectangle becoming a waterlogged interruption area.

[0138] S10352. Obtain, based on the average grayscale value of the current first minimum circumscribed rectangle and the sum of the average grayscale values, a second possible value for the interruption area corresponding to the current first minimum circumscribed rectangle to become a waterlogged interruption area.

[0139] S10353. Obtain a final possible value of the interruption area corresponding to the current first minimum circumscribed rectangle becoming a waterlogged interruption area according to the first possible value and the second possible value.

[0140] S10354. When the final possible value is greater than a third preset threshold, determine that the interruption area corresponding to the current first minimum circumscribed rectangle is a waterlogged interruption area.

[0141] The waterlogging interruption area can be obtained by the following formula:

[0142]

[0143] In this formula, Indicates the possibility that the first minimum enclosing rectangle is the water interruption area, represents the normalization function, represents the area of ​​the first minimum circumscribed rectangle, represents the sum of the areas of all first minimum enclosing rectangles, represents the difference between the area of ​​the first minimum enclosing rectangle and the areas of all first minimum enclosing rectangles, and It can be said that the first minimum enclosing rectangle is the first possible value of the water interruption area. The larger the value is, the greater the possibility that the interruption area corresponding to the first minimum circumscribed rectangle is a waterlogging interruption area. Represents the average grayscale value of all pixels in the first minimum bounding rectangle. Represents the sum of the average grayscale values ​​of all first minimum bounding rectangles, represents the difference between the average grayscale value of all pixels in the first minimum bounding rectangle and the sum of the average grayscale values ​​of all first minimum bounding rectangles, and Indicates that the first minimum bounding rectangle is the second possible value of the water interruption area, The larger the value, the greater the possibility that the interruption area corresponding to the first minimum circumscribed rectangle is a waterlogged interruption area. Then combine them, The larger the value is, the more likely it is that the first minimum circumscribed rectangle is a waterlogging interruption area.

[0144] The third preset threshold is used to screen whether the interruption area corresponding to the first minimum circumscribed rectangle is a water-logged interruption area. The value range of the third preset threshold can be determined to be (0,1) through the range of possible values ​​of the first minimum circumscribed rectangle being a water-logged interruption area. When the value of the third preset threshold is closer to 0, it means that the standard for using the third preset threshold to screen whether the interruption area corresponding to the first minimum circumscribed rectangle is a water-logged interruption area is looser. When the value of the third preset threshold is closer to 1, it means that the standard for using the third preset threshold to screen whether the interruption area corresponding to the first minimum circumscribed rectangle is a water-logged interruption area is stricter. This embodiment gives a specific example, that is, this embodiment requires a stricter screening standard to determine the candidate crack end pixel points. For example, the third preset threshold is set to 0.7, then When the value is greater than 0.7, the interruption area corresponding to the first minimum circumscribed rectangle is the waterlogging interruption area.

[0145] In this way, all waterlogging interruption areas can be extracted and marked.

[0146] S104, repairing the cracks in the water accumulation interruption area to obtain complete cracks corresponding to the water accumulation interruption area.

[0147] Specifically, for each water interruption area, the last pixel point corresponding to the water interruption area and the pixel point at the end of the target crack are obtained, and the last pixel point in each water interruption area and the pixel point at the end of the target crack are connected to obtain a complete crack corresponding to each water interruption area.

[0148] Based on the above operations, a possibility analysis is performed on the waterlogging interruption area based on its characteristics to determine the specific location of the waterlogging interruption area. This step will continue to repair the interruption area based on the possibility analysis results and the location of the waterlogging interruption area obtained in the above steps.

[0149] The water interruption area basically presents a coherent crack state before the water accumulates. Since the reflection of the water interrupts part of the crack area, there is a certain directional trend relationship between the interrupted crack area and the uninterrupted crack areas at both ends. This relationship can be judged by the directional trend of the pixels at both ends of the interrupted area. Here, by establishing a line between the two end points of the water interruption area, the cracks in the water interruption area can be repaired.

[0150] S105. Classify the road surface quality according to the repair result.

[0151] Determine the distance from other pixels in the minimum circumscribed rectangle of the water interruption area to the straight line to analyze the possibility of it being a crack in the interruption area. Figure 8As shown, the specific operations are as follows: Steps S1051-S1053:

[0152] S1051. Obtain the entire crack area according to the complete cracks corresponding to all water interruption areas and all crack growth areas.

[0153] S1052. Establish a second minimum circumscribed rectangle for the entire crack area.

[0154] A second minimum circumscribed rectangle is established for the entire crack area, and the area t of the second minimum circumscribed rectangle is obtained.

[0155] S1053: Classify the road surface quality of the road according to the pixel ratio of the crack area in the second minimum circumscribed rectangle and the area of ​​the second minimum circumscribed rectangle.

[0156] Get the number of all pixels in the second minimum circumscribed rectangle and the number of pixels in the crack area in the second minimum circumscribed rectangle, calculate the ratio of the number of pixels in the crack area in the second minimum circumscribed rectangle to the number of all pixels in the second minimum circumscribed rectangle, to get the pixel ratio of the crack area in the second minimum circumscribed rectangle.

[0157] Specifically, the product of the pixel ratio of the crack area in the second minimum circumscribed rectangle and the area of ​​the second minimum circumscribed rectangle is obtained; when it is detected that the product is less than the first threshold, the degree of pavement damage of the road is determined to be that the road surface begins to settle; when it is detected that the product is less than or equal to the second threshold and greater than or equal to the first threshold, the degree of pavement damage of the road is determined to be that settlement has occurred; when it is detected that the product is greater than the second threshold, the degree of pavement damage of the road is determined to be that it is about to collapse.

[0158] The road quality can be expressed by the following formula:

[0159] ;

[0160] In this formula, Indicates the road quality. represents the area of ​​the second smallest enclosing rectangle, Represents the pixel percentage of the crack area in the second smallest circumscribed rectangle, where The larger the value, the larger the crack area and the greater the degree of damage to the road surface. Represents the normalization function.

[0161] For example, the first threshold may be 0.3, and the second threshold may be 0.6.

[0162] like If the value is less than 0.3, the road surface is considered to have begun to settle. Control measures: Use asphalt or concrete materials to level the settlement area and restore the road surface flatness. Dig up the settlement area, add sand and gravel or other stable materials, and then re-pave the road surface after compaction. Check the drainage facilities around the settlement area to ensure smooth drainage and prevent water accumulation from exacerbating settlement. Regularly check the settlement area, monitor the settlement trend, and repair small cracks or depressions in a timely manner.

[0163] like If the value is between [0.3,0.6], the road surface is considered to have experienced moderate settlement. Control measures: Dig out the soft foundation soil in the settlement area, and refill and compact it with more stable materials, such as lime soil or cement stabilized soil. Use grouting technology (such as cement slurry or resin slurry) to fill cracks and voids in the settlement area to enhance the foundation strength. Add drainage ditches or drainage pipes to reduce the erosion of surface and groundwater on the roadbed. Re-pave the road surface in the settlement area and use anti-settlement materials (such as high-strength asphalt or reinforced concrete) to improve the bearing capacity.

[0164] If v is greater than 0.6, the road surface is considered to be about to collapse. Control measures: Dig out the soft soil layer in the settlement area, replace the deep layer with high-strength materials (such as crushed stone, gravel) and compact it layer by layer. Use new foundation reinforcement technologies (such as CFG piles, dynamic tamping or reinforced soil technology) to improve the bearing capacity of the foundation. If the settlement is caused by underground cavities or pipe leakage, the damaged pipes should be repaired or replaced first, and then the cavities should be filled and compacted. Build underground drainage networks, rainwater collection pools or infiltration and drainage systems to ensure rapid drainage of water and reduce foundation erosion. Install settlement monitoring equipment to monitor foundation changes in real time, and establish an early warning system to prevent further settlement.

[0165] The present disclosure is mainly aimed at the problems existing in the quality detection under the background of road monitoring. During the detection, factors such as water accumulation often occur, which causes the detection results to be blocked, thereby affecting the judgment of the road quality detection results. The present disclosure studies the problem that cracks caused by road surface settlement due to water accumulation are interrupted in the edge detection results, and repairs the cracks in the interrupted area. This method improves the efficiency of road detection and promotes the further development of road quality detection technology. Furthermore, the settlement area where water accumulation appears on the road will form a reflective area. Then this reflective area appears as a higher gray value area on the computer vision edge detection image, and this area interrupts the cracks in the settlement area in the edge detection. The present disclosure mainly locates the interrupted area according to its characteristics, and repairs the interrupted area according to the extension direction of the pixel point after determining the position. The present disclosure solves the impact of road water accumulation on road settlement characteristics that cannot be avoided based on edge detection technology, and improves the accuracy of road quality detection to a certain extent.

[0166] Fig. 9A structural block diagram of a municipal road construction safety monitoring device provided by an embodiment of the present invention is shown in FIG. Fig. 9 As shown, the municipal road construction safety monitoring device includes:

[0167] The first acquisition module 11 is used to acquire an image of a road, pre-process the image of the road, and acquire a road surface settlement image.

[0168] The second acquisition module 12 acquires the interruption area in the road subsidence image.

[0169] The third acquisition module 13 acquires the waterlogged interruption area in the interruption area.

[0170] The fourth acquisition module 14 repairs the cracks in the water accumulation interruption area to obtain complete cracks corresponding to the water accumulation interruption area.

[0171] The fifth acquisition module 15 classifies the road surface quality of the road according to the repair result.

[0172] In one embodiment, the first acquisition module 11 is specifically configured to:

[0173] Acquire a subsidence area image in the image of the road;

[0174] grayscale the subsidence area image to obtain a subsidence area grayscale image;

[0175] Performing edge detection on the grayscale image of the sedimentation area to obtain an edge detection image;

[0176] The road surface subsidence image in the subsidence area image is acquired according to the edge detection image.

[0177] In one embodiment, in the aspect of acquiring the interruption area in the road subsidence image, the second acquisition module 12 is specifically used to:

[0178] Extracting crack pixel points corresponding to the crack area in the road surface settlement image;

[0179] Performing region growth on the crack pixel points to obtain multiple crack growth regions;

[0180] The following steps are performed for each of the crack growth regions:

[0181] In the other crack growth regions, according to the last pixel point of the crack interruption end in the current crack growth region and the crack end pixel points in the other crack growth regions, a candidate crack growth region belonging to the same crack as the current crack growth region is obtained;

[0182] In the candidate crack growth region, according to the crack width of the current crack growth region and the crack width of the candidate crack growth region, a target crack growth region belonging to the same crack as the current crack growth region is obtained;

[0183] The portion between the current crack growth region and the target crack growth region is the interruption region.

[0184] In one embodiment, in the other crack growth regions, in terms of acquiring candidate crack growth regions belonging to the same crack as the current crack growth region according to the last pixel point of the crack interruption end in the current crack growth region and the crack end pixel points of the other crack growth regions, the second acquisition module 12 is specifically used to:

[0185] Obtaining the last pixel point of the crack break end in the current crack growth area;

[0186] Get the previous pixel point adjacent to the last pixel point;

[0187] Obtaining the angle between a pixel vector and each candidate vector, wherein the pixel vector is a vector formed by the previous pixel and the last pixel, and each candidate vector is a vector formed by the last pixel and each other pixel of the crack end of the crack growth area;

[0188] Obtaining the modulus length of each of the candidate vectors;

[0189] In each of the included angles and the corresponding modulus lengths, a candidate crack end pixel point that meets a first preset threshold is obtained;

[0190] Among them, the crack growth region corresponding to the pixel point at the end of the candidate crack is the candidate crack growth region.

[0191] In one embodiment, in the candidate crack growth region, in terms of acquiring a target crack growth region belonging to the same crack as the current crack growth region according to the crack width of the current crack growth region and the crack width of the candidate crack growth region, the second acquisition module 12 is specifically used to:

[0192] Calculating a first average value of crack widths within a preset length from the last pixel point in the current crack growth region;

[0193] A second average value of crack width within the preset length from the pixel point at the end of the candidate crack in each of the candidate crack growth regions is calculated.

[0194] According to the first average value and each of the second average values, a target crack end pixel point that meets a second preset threshold is obtained from the candidate crack end pixel points, and the candidate crack growth region where the target crack end pixel point is located is the target crack growth region.

[0195] In one embodiment, in the aspect of obtaining the waterlogging interruption area in the interruption area, the third obtaining module 13 is specifically used for:

[0196] Establishing a first minimum bounding rectangle for each of the interruption areas, and obtaining the area of ​​each of the first minimum bounding rectangles;

[0197] Obtain the sum of the areas of all the first minimum circumscribed rectangles;

[0198] Calculate the average grayscale value of all pixels in each of the first minimum circumscribed rectangles;

[0199] Obtaining the sum of average grayscale values ​​of all the first minimum bounding rectangles;

[0200] The water accumulation interruption area in the interruption area is obtained according to the area of ​​each of the first minimum circumscribed rectangles, the sum of the areas, the average grayscale value of each of the first minimum circumscribed rectangles, and the sum of the average grayscale values.

[0201] In one embodiment, in acquiring the waterlogged interruption area in the interruption area according to the area of ​​each of the first minimum circumscribed rectangles, the sum of the areas, the average grayscale value of each of the first minimum circumscribed rectangles, and the sum of the average grayscale values, the third acquisition module 13 is specifically used to:

[0202] For each of the first minimum bounding rectangles, perform the following steps:

[0203] According to the area of ​​the current first minimum circumscribed rectangle and the sum of the areas, obtaining a first possible value of the interruption area corresponding to the current first minimum circumscribed rectangle becoming the waterlogged interruption area;

[0204] According to the average grayscale value of the current first minimum circumscribed rectangle and the sum of the average grayscale values, obtaining a second possible value of the interruption area corresponding to the current first minimum circumscribed rectangle becoming the waterlogged interruption area;

[0205] Obtaining a final possible value for the interruption area corresponding to the current first minimum circumscribed rectangle to become the waterlogged interruption area according to the first possible value and the second possible value;

[0206] When the final possible value is greater than a third preset threshold, it is determined that the interruption area corresponding to the current first minimum circumscribed rectangle is the water accumulation interruption area.

[0207] In one embodiment, in the aspect of repairing the cracks in the water accumulation interruption area to obtain the complete cracks corresponding to the water accumulation interruption area, the fourth acquisition module 14 is specifically used to:

[0208] For each of the water accumulation interruption areas, obtaining the last pixel point corresponding to the water accumulation interruption area and the target crack end pixel point;

[0209] The last pixel point in each of the water accumulation interruption areas is connected to the pixel point at the end of the target crack to obtain a complete crack corresponding to each of the water accumulation interruption areas.

[0210] In one embodiment, in the aspect of classifying the road surface quality of the road according to the repair result, the fifth acquisition module 15 is specifically used to include:

[0211] Obtaining the entire crack region according to the complete cracks corresponding to all the water interruption regions and all the crack growth regions;

[0212] Establishing a second minimum circumscribed rectangle for the entire crack region;

[0213] The road surface quality of the road is classified according to the pixel ratio of the crack area in the second minimum circumscribed rectangle and the area of ​​the second minimum circumscribed rectangle.

[0214] In one embodiment, in the aspect of classifying the road surface quality of the road according to the pixel point ratio of the crack area in the second minimum circumscribed rectangle and the area of ​​the second minimum circumscribed rectangle, the fifth acquisition module 15 is specifically used to:

[0215] Obtaining the product of the pixel ratio of the crack area in the second minimum circumscribed rectangle and the area of ​​the second minimum circumscribed rectangle;

[0216] When it is detected that the product is less than a first threshold, determining that the road surface damage degree of the road is that the road surface begins to settle;

[0217] When it is detected that the product is less than or equal to the second threshold and greater than or equal to the first threshold, determining that the road surface damage degree of the road is that settlement has occurred;

[0218] When it is detected that the product is greater than the second threshold, it is determined that the road surface damage degree of the road is about to collapse.

[0219] It should be noted that the sequence of the above embodiments of the present invention is for description only and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0220] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A municipal road construction safety monitoring method, characterized in that: The method comprises: Acquire an image of a road, pre-process the image of the road, and acquire a road surface settlement image; Acquire a discontinuity region in the road surface settlement image; Acquire a waterlogging interruption area in the interruption area; Repairing the cracks in the water accumulation interruption area to obtain complete cracks corresponding to the water accumulation interruption area; classifying the road surface quality of the road according to the restoration results; The obtaining of the interrupted area in the road subsidence image includes: Extracting crack pixel points corresponding to the crack area in the pavement settlement image; performing regional growth on the crack pixel points based on the gray value difference between the crack pixel points and the crack pixel points in the neighborhood to obtain multiple crack growth areas; The following steps are performed for each of the crack growth regions: In the other crack growth regions, according to the last pixel point of the crack interruption end in the current crack growth region and the crack end pixel points in the other crack growth regions, a candidate crack growth region belonging to the same crack as the current crack growth region is obtained, including: Obtain the last pixel point of the crack interruption end in the current crack growth region; obtain the previous pixel point adjacent to the last pixel point; obtain the angle between the pixel point vector and each candidate vector, the pixel point vector is a vector formed by the previous pixel point and the last pixel point, and each candidate vector is a vector formed by the last pixel point and each other crack end pixel point in the crack growth region; obtain the modulus of each candidate vector; among each angle and the corresponding modulus, obtain the candidate crack end pixel point that meets the first preset threshold; wherein the crack growth region corresponding to the candidate crack end pixel point is the candidate crack growth region; In the candidate crack growth region, according to the crack width of the current crack growth region and the crack width of the candidate crack growth region, obtaining a target crack growth region belonging to the same crack as the current crack growth region, comprising: Calculate the first average value of the crack width within a preset length from the last pixel point under the straight line where the crack direction is located in the current crack growth area; calculate the second average value of the crack width within the preset length from the candidate crack end pixel point under the straight line where the crack direction is located in each candidate crack growth area; wherein the straight line where the crack direction is located refers to the line connecting the crack end pixel points in the crack growth area; according to the first average value and each of the second average values, obtain a target crack end pixel point that meets a second preset threshold from the candidate crack end pixel points, and the candidate crack growth area where the target crack end pixel point is located is the target crack growth area; wherein the value range of the second preset threshold is (0,1); The portion between the current crack growth region and the target crack growth region is the interruption region.

2. A municipal road construction safety monitoring method according to claim 1, characterized in that: The preprocessing of the road image to obtain a road surface settlement image includes: Acquire a subsidence area image in the image of the road; grayscale the subsidence area image to obtain a subsidence area grayscale image; Performing edge detection on the grayscale image of the sedimentation area to obtain an edge detection image; The road surface subsidence image in the subsidence area image is acquired according to the edge detection image.

3. A municipal road construction safety monitoring method according to claim 1, characterized in that: The obtaining of the waterlogged interruption area in the interruption area comprises: Establishing a first minimum bounding rectangle for each of the interruption areas, and obtaining the area of ​​each of the first minimum bounding rectangles; Obtain the sum of the areas of all the first minimum circumscribed rectangles; Calculate the average grayscale value of all pixels in each of the first minimum circumscribed rectangles; Obtaining the sum of average grayscale values ​​of all the first minimum bounding rectangles; The water accumulation interruption area in the interruption area is obtained according to the area of ​​each of the first minimum circumscribed rectangles, the sum of the areas, the average grayscale value of each of the first minimum circumscribed rectangles, and the sum of the average grayscale values.

4. A municipal road construction safety monitoring method according to claim 3, characterized in that: The acquiring the waterlogged interruption area in the interruption area according to the area of ​​each of the first minimum circumscribed rectangles, the sum of the areas, the average grayscale value of each of the first minimum circumscribed rectangles, and the sum of the average grayscale values ​​comprises: For each of the first minimum bounding rectangles, perform the following steps: According to the area of ​​the current first minimum circumscribed rectangle and the sum of the areas, the first possible value of the interruption area corresponding to the current first minimum circumscribed rectangle becoming the waterlogged interruption area is obtained, and the first possible value is expressed as ,in represents the area of ​​the first minimum circumscribed rectangle, represents the sum of the areas of all first minimum circumscribed rectangles; According to the average grayscale value of the current first minimum circumscribed rectangle and the sum of the average grayscale values, the interruption area corresponding to the current first minimum circumscribed rectangle is obtained as a second possible value of the waterlogged interruption area, and the second possible value is expressed as ,in Represents the average grayscale value of all pixels in the first minimum bounding rectangle. Represents the sum of the average gray values ​​of all first minimum bounding rectangles; Obtaining a final possible value for the interruption area corresponding to the current first minimum circumscribed rectangle to become the waterlogged interruption area according to the first possible value and the second possible value; When the final possible value is greater than a third preset threshold, the interruption area corresponding to the current first minimum circumscribed rectangle is determined to be the waterlogged interruption area; the value range of the third preset threshold is (0, 1).

5. A municipal road construction safety monitoring method according to claim 4, characterized in that: The repairing of the cracks in the water accumulation interruption area to obtain a complete crack corresponding to the water accumulation interruption area includes: For each of the water accumulation interruption areas, obtaining the last pixel point corresponding to the water accumulation interruption area and the target crack end pixel point; The last pixel point in each of the water accumulation interruption areas is connected to the pixel point at the end of the target crack to obtain a complete crack corresponding to each of the water accumulation interruption areas.

6. A municipal road construction safety monitoring method according to claim 1, characterized in that: The road surface quality of the road is classified according to the repair result, including: Obtaining the entire crack region according to the complete cracks corresponding to all the water interruption regions and all the crack growth regions; Establishing a second minimum circumscribed rectangle for the entire crack region; The road surface quality of the road is classified according to the pixel ratio of the crack area in the second minimum circumscribed rectangle and the area of ​​the second minimum circumscribed rectangle.

7. A municipal road construction safety monitoring method according to claim 6, characterized in that: The classifying the road surface quality of the road according to the pixel point ratio of the crack area in the second minimum circumscribed rectangle and the area of ​​the second minimum circumscribed rectangle includes: Obtaining the product of the pixel ratio of the crack area in the second minimum circumscribed rectangle and the area of ​​the second minimum circumscribed rectangle; When it is detected that the product is less than a first threshold, determining that the road surface damage degree of the road is that the road surface begins to settle; When it is detected that the product is less than or equal to the second threshold and greater than or equal to the first threshold, determining that the road surface damage degree of the road is that settlement has occurred; When it is detected that the product is greater than the second threshold, it is determined that the road surface damage degree of the road is about to collapse.

Images