A waterproof coating spraying control method and system for improving construction adaptability

Abstract

The present application relates to the field of image processing technology, and specifically to a waterproof coating spraying control method and system for improving construction adaptability, the method comprising: real-time acquisition of a coating grayscale image during the waterproof coating spraying process; acquisition of highlighted pixels; clustering all highlighted pixels in the coating grayscale image, and acquiring the highlight reflection coefficient of each pixel based on the size and grayscale characteristics of the cluster in which each pixel is located, as well as the distance from the remaining clusters; acquiring the reflective reflection coefficient of each pixel based on the shape characteristics, branch characteristics, and deformation characteristics of the edge where each pixel is located, and acquiring the reflective interference authenticity of each pixel in combination with the grayscale characteristics and highlight reflection coefficient within the neighborhood of the edge where each pixel is located, and then correcting the preset initial weight of each pixel. The present application improves the quality of the filtered coating image by adaptively calculating the weight of each pixel, thereby more accurately judging the coating defect situation.

Description

Technical Field

[0001] The present application relates to the field of image processing technology, and in particular to a waterproof coating spraying control method and system for improving construction adaptability. Background Art

[0002] Currently, leak-proofing renovations in older residential communities typically involve cleaning and repairing the roof base, then spraying a large area with waterproof coating to improve the waterproofing performance of older buildings. Currently, most automated spraying equipment or robots use process parameter adjustments to monitor system parameters during the spraying process, combined with coating quality test results, to assess the health of the spraying system and determine whether the spraying equipment parameters need adjustment. During this process, visually assisted methods are primarily used for coating quality inspection, capturing and processing image data to determine if the coating has quality defects.

[0003] However, the rooftop conditions of houses in older communities are relatively complex. When spraying waterproofing materials, visual inspection of coating quality may be affected by environmental factors (such as reflections, shadows, and dim weather). This can result in a lot of noise in the collected image data, poor image quality, and an inability to accurately judge the quality of the coating. Therefore, filtering algorithms are often used to optimize image quality. The traditional non-local mean filtering algorithm (NLM) takes into account the non-local statistical self-similarity of the image. It calculates the similarity between non-local pixels to determine weights and then restores the image by weighted averaging. However, during the spraying process, the coating surface will produce reflections due to lighting factors, which in turn affects the self-similarity of the image and distorts the original coating features. The final filtered image quality is still poor, making it difficult to further determine coating defects. Summary of the Invention

[0004] In order to solve the above technical problems, the purpose of this application is to provide a waterproof coating spraying control method and system that improves construction adaptability. The technical solutions adopted are as follows:

[0005] In a first aspect, an embodiment of the present application provides a method for controlling the spraying of a waterproof coating to improve construction adaptability, the method comprising the following steps:

[0006] Real-time acquisition of grayscale images of the waterproof coating during spraying;

[0007] Comparing the grayscale value of each pixel with a preset segmentation threshold to obtain highlighted pixels; clustering all highlighted pixels in the current coating grayscale image, and obtaining neighboring clusters of each cluster based on the distance between each cluster and other clusters; obtaining the highlight reflectivity coefficient of each pixel based on the number of pixels in the cluster where each pixel is located, the average level of distances between the cluster and the remaining clusters, and the difference between the average grayscale value of the cluster where the pixel is located and the average grayscale value of its neighboring clusters;

[0008] Obtain the edges and deformed pixels in the current coating grayscale image, and obtain the reflection coefficient of each pixel based on the similarity between the edge where each pixel is located and other edges, the number of endpoints on the edge, and the number of deformed pixels in the horizontal and vertical directions of each deformed pixel in the edge;

[0009] According to the grayscale category in the neighborhood of the edge where each pixel is located, and the difference between the average grayscale level in the neighborhood of each pixel and the average grayscale level of the coating grayscale image where it is located, the reflection coverage authenticity of each pixel is obtained, and the reflection interference authenticity of each pixel is obtained by combining the highlight reflection coefficient and reflection reflection coefficient of each pixel, and then the preset initial weight of each pixel in the coating grayscale image at the current moment is corrected.

[0010] Preferably, the highlighted pixel points refer to pixel points whose grayscale values ​​are greater than or equal to a preset segmentation threshold.

[0011] Preferably, the process of obtaining the neighboring clusters of each cluster is: calculating the Euclidean distance between the center points of a single cluster and all other clusters, and taking the clusters with Euclidean distances less than a preset neighboring threshold as the neighboring clusters of the cluster.

[0012] Preferably, the calculation formula for the highlight reflection coefficient of each pixel is: Where, is the highlight reflection coefficient of the i-th pixel, is the total number of pixels contained in the cluster where the i-th pixel is located, is the mean of the Euclidean distances between the cluster where the i-th pixel is located and the center points of the remaining clusters, is the grayscale mean of the pixel in the cluster where the i-th pixel is located, is the grayscale mean of the pixels in the jth neighboring cluster of the cluster where the i-th pixel is located, and J is the total number of neighboring clusters of the cluster where the i-th pixel is located.

[0013] Preferably, the process of acquiring the deformed pixel points is: acquiring the curvature of each pixel point in each edge of the coating grayscale image at the current moment, and taking the pixel points whose curvature is greater than or equal to the preset deformation threshold as the deformed pixel points of the coating grayscale image at the current moment.

[0014] Preferably, the calculation formula for the reflection coefficient of each pixel point is: Where, is the reflection coefficient of the i-th pixel, is the mean of the cosine similarity between the edge where the i-th pixel is located and all other edges, is the total number of end edge points on the edge where the i-th pixel is located, is the total number of deformed pixels in the horizontal and vertical directions of the x-th deformed pixel in the edge where the i-th pixel is located. is the total number of deformed pixels in the edge where the i-th pixel is located; wherein, the end edge point on the edge refers to a pixel point that has only two edge pixels in the eight neighborhoods of the edge.

[0015] Preferably, the calculation formula for the reflection coverage authenticity of each pixel point is: Where, is the reflection coverage authenticity of the i-th pixel, is the total number of growth areas of neighboring pixels on the edge of the i-th pixel, is the grayscale mean of all pixels in the neighborhood window of the i-th pixel, is the grayscale mean of the grayscale image of the coating where the i-th pixel is located; wherein the neighborhood window of the i-th pixel refers to a window with the i-th pixel as the center and a preset distance as the radius; the neighboring pixels of the edge refer to all pixels within the neighborhood window of all pixels in the edge; the growth area of ​​the neighboring pixels of the edge refers to a plurality of growth areas into which the neighboring pixel areas of each edge are divided using a region growing algorithm.

[0016] Preferably, the reflection interference authenticity of each pixel point refers to the forward fusion result of the reflection coverage authenticity, highlight reflection coefficient and reflection reflection coefficient of each pixel point.

[0017] Preferably, the preset initial weight of each pixel in the coating grayscale image at the current moment is corrected, and the calculation formula of the corrected weight of each pixel is: Where, is the corrected weight of the i-th pixel, To preset the initial weight, is the reflection interference authenticity of the i-th pixel, and norm() is the normalization function.

[0018] In the second aspect, an embodiment of the present application also provides a waterproof coating spraying control system for improving construction adaptability, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any one of the above-mentioned waterproof coating spraying control methods for improving construction adaptability.

[0019] This application has at least the following beneficial effects:

[0020] This application analyzes the interference characteristics of reflective highlights and reflective reflections in the coating images collected during the waterproof coating spraying process, constructs highlight reflection coefficients and reflective reflection coefficients, which are used to distinguish whether the reflective characteristics of pixel points are caused by light or coating defects, providing a more accurate basis for calculating pixel weights in subsequent image filtering; further analyzes the interference characteristics of reflections and shadows, constructs the reflection coverage authenticity, evaluates the actual situation of each pixel point in the coating grayscale image being interfered with by reflections, and then corrects the preset initial weight of each pixel point obtained by the NLM algorithm, solving the problem that the image features contained in each pixel point are less accurate due to the interference of reflections, and the use of traditional NLM algorithms to calculate weights will cause distortion of the original image features, resulting in poor quality of the filtered image, thereby improving the image quality in the coating quality detection process, and thus being able to more accurately judge coating defects. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present application or the prior art, the following is a brief introduction to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0022] Figure 1 A flowchart of a method for controlling the spraying of a waterproof coating to improve construction adaptability provided in one embodiment of the present application;

[0023] Figure 2 A flowchart for obtaining the authenticity of reflective interference of each pixel point provided in one embodiment of the present application. DETAILED DESCRIPTION

[0024] To further illustrate the technical means and effects employed by this application to achieve the intended invention objectives, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a waterproof coating spray control method and system for improving construction adaptability proposed in this application. In the following description, different references to "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0025] 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 application belongs.

[0026] The following describes in detail a specific solution of a waterproof coating spraying control method and system for improving construction adaptability provided by the present application in conjunction with the accompanying drawings.

[0027] See also Figure 1 , which shows a flowchart of a method for controlling the spraying of a waterproof coating for improving construction adaptability provided by an embodiment of the present application, the method comprising the following steps:

[0028] Step 1: Acquire the grayscale image of the waterproof coating during the spraying process in real time.

[0029] The automated spraying equipment uses a CMOS camera installed on either side of the spray gun. As the spray gun moves, the camera captures images of the sprayed coating in the direction opposite to its movement. (For example, if the spray gun is moving eastward, the camera on the west side captures data. If the spray gun switches from east-west to north-south movement, the camera can be rotated to switch from east-west to north-south.) RGB image data of the coating is captured every T seconds, and the grayscale averaging method is used to convert the RGB image into grayscale image data. In this embodiment, T = 0.2.

[0030] Step 2: Compare the grayscale value of each pixel with the preset segmentation threshold to obtain the highlighted pixel points; cluster all the highlighted pixel points in the grayscale image of the coating at the current moment, and obtain the adjacent clusters of each cluster based on the distance between each cluster and other clusters; obtain the highlight reflection coefficient of each pixel point based on the number of pixels in the cluster where each pixel point is located, the average level of the distance between it and the remaining clusters, and the difference between the average grayscale value of the cluster where it is located and the average grayscale value of its adjacent clusters.

[0031] Typically, evaluation criteria for the spray quality of waterproof coatings include, but are not limited to, cracking, leaks, and uneven thickness. These conditions are primarily manifested in the coating image as polymorphic edges in localized areas, leaks of the roof base color, and convex bulges and indentations. For older residential rooftops, cleaning and repairing them will result in a relatively smooth coating. When the waterproof coating is sprayed, sufficient light will cause reflections on the relatively smooth coating surface. There are two main types of reflective phenomena. One is highlight, which is when a local bright area or bright point appears in the image. The distribution is relatively concentrated and there are certain diffusion characteristics (for example, when a highlight appears in a certain area, the brightness of the local area is extremely high, and many bright spots will appear around the area. The distribution density of these bright spots will gradually decrease, and the brightness will also decrease accordingly). The second type of reflective phenomenon is reflection, which is when the coating surface is relatively smooth and thus mirror reflection occurs in a well-lit environment. However, since the coating is not as smooth as a mirror, the clarity of the reflection is relatively low and it will produce certain deformations. In the image, it mainly appears as edges of different shapes and color changes on the coating surface.

[0032] For the convex hull formed by the area of ​​uneven thickness in the layer, it may also appear as a separate bright spot in the image under illumination, which is similar to the first reflection situation. In addition, due to the uneven height of houses in old communities, the second reflection situation on the roof of the house is usually the reflection of nearby buildings, railings, and self-built facilities. After the various reflections overlap, they mainly appear as polymorphic edges in the image, which are similar to the characteristics when defects occur in the coating. Therefore, it is necessary to make a certain distinction to prevent the reflection phenomenon from being misjudged as a coating defect. Specifically, when there is a convex hull highlight in the grayscale image of the coating, compared with the highlight area formed by the reflection, the distribution of the convex hull highlight area is more discrete and more isolated, and compared with the highlight area formed by the reflection, the brightness of the convex hull highlight area is more prominent.

[0033] In order to characterize the features of the highlight area in the image, the grayscale value of each pixel in the coating grayscale image obtained at the current moment is used as input, and the cross-validation method is used to output the segmentation threshold of the grayscale value. The segmentation threshold is recorded as the preset segmentation threshold. When the grayscale value of a certain pixel is greater than or equal to the preset segmentation threshold, the pixel is recorded as a highlight pixel. For the coating grayscale image collected at the current moment, the coordinates of all the highlight pixels are used as input, and the DPC density clustering algorithm is used. The cutoff distance is n, and n is 3 in the present embodiment, and the output is a plurality of clusters. For a single cluster, the Euclidean distance between it and the center point of all the remaining clusters is calculated, and the cluster whose Euclidean distance is less than the preset proximity threshold is used as the adjacent cluster of the cluster. In the present embodiment, the preset proximity threshold is 12, and the implementer can determine it according to the actual situation. The DPC density clustering algorithm is a well-known technology, and the specific process is not repeated here.

[0034] The fewer pixels a single cluster contains, the larger the distance between it and the other clusters, and the brighter the mean grayscale value of the pixels within the cluster is compared to the brightness of the adjacent clusters, the more it indicates that the pixels contained in the cluster correspond to the highlights caused by the convex hull defects of the coating itself, rather than the highlights caused by reflections.

[0035] As a preferred embodiment, the highlight reflection coefficient of each pixel point is obtained based on the number of pixels in the cluster where each pixel point is located, the average level of the distance between the cluster and the remaining clusters, and the difference between the average grayscale value of the cluster where the pixel point is located and the average grayscale value of its adjacent clusters, which is used to characterize the possibility that a single pixel point is a pixel point in the highlight area generated by reflection.

[0036] In this embodiment, the highlight reflection coefficient of the i-th pixel is recorded as , its specific expression is: Where, is the highlight reflection coefficient of the i-th pixel, is the total number of pixels contained in the cluster where the i-th pixel is located, is the mean of the Euclidean distances between the cluster where the i-th pixel is located and the center points of the remaining clusters, is the grayscale mean of the pixel in the cluster where the i-th pixel is located, is the mean grayscale value of the jth neighboring cluster of the cluster where the i-th pixel is located, and J is the total number of neighboring clusters of the cluster where the i-th pixel is located. It should be noted that when the i-th pixel is not a highlight pixel, the highlight reflection coefficient of the i-th pixel is set to 1.

[0037] The larger the value of , the more likely the i-th pixel is to be a pixel in the highlight area caused by reflection.

[0038] Step 3: Obtain the edges and deformed pixels in the coating grayscale image at the current moment, and obtain the reflection coefficient of each pixel based on the similarity between the edge where each pixel is located and other edges, the number of endpoints on the edge, and the number of deformed pixels in the horizontal and vertical directions of each deformed pixel in the edge.

[0039] Furthermore, regarding the reflections produced by reflections, since the roofs of houses in old residential areas are mainly surrounded by other buildings, railings, self-built facilities, etc., the edges of the reflections produced are relatively flat, and deformation will only occur in local areas, and the areas where deformation occurs are relatively consistent in spatial position. In addition, since they are reflections of surrounding buildings, the colors of these edges are more complex and varied, and are more different from the colors of solid-color coatings. Compared to the edges formed by reflections, the edges produced when cracks appear on the coating are obvious multi-branched edges. The larger the cracks, the more similar the shapes of the two edges will be in the edge detection results.

[0040] Based on the above analysis, the coating grayscale image at the current moment is used as input, and the Canny edge detection algorithm and the expansion and corrosion algorithm are used to obtain all edges in the coating grayscale image. The curvature of each pixel in each edge of the coating grayscale image at the current moment is calculated, and all curvatures are used as inputs of the Otsu threshold method to obtain a segmentation threshold. The segmentation threshold is recorded as the preset deformation threshold, and the pixels with a curvature greater than or equal to the preset deformation threshold are used as the deformed pixels of the coating grayscale image at the current moment; when there are more deformed pixels on the horizontal and vertical axes of the coordinate point where a single deformed pixel is located, it means that the edge where the deformed pixel is located is more likely to be the edge of the reflection caused by the reflection of the coating surface. In addition, the more branches there are on the edge, the more endpoints there are on the edge. In order to characterize the edge branch characteristics generated when the coating itself cracks, the pixel point with only two edge pixels in the eight neighborhoods of a single edge is used as the end edge point of the edge, and the end edge point is used to characterize the endpoint of the edge. It should be noted that the Canny edge detection algorithm, the expansion and corrosion algorithm, and the Otsu threshold method are all well-known technologies, and the specific process will not be repeated here.

[0041] As a preferred embodiment, the reflection coefficient of each pixel point is obtained based on the similarity between the edge where each pixel point is located and other edges, the number of endpoints on the edge, and the number of deformed pixels in the horizontal and vertical directions of each deformed pixel point in the edge, which is used to characterize the possibility that a single pixel point is a pixel point in the edge generated by the reflection of the coating surface.

[0042] In this embodiment, the reflection coefficient of the i-th pixel is recorded as , its specific expression is: Where, is the reflection coefficient of the i-th pixel, is the mean of the cosine similarity between the edge where the i-th pixel is located and all other edges, is the total number of end edge points on the edge where the i-th pixel is located, is the total number of deformed pixels in the horizontal and vertical directions of the x-th deformed pixel in the edge where the i-th pixel is located. is the total number of deformed pixels on the edge where the i-th pixel is located. It should be noted that if the i-th pixel is not on the edge, the reflection coefficient of the i-th pixel is set to 1.

[0043] The meaning of this relationship is: when the similarity between the edge where the i-th pixel point is located and the rest of the edges is lower, and the number of edge points on the upper end of the edge is smaller, and the number of deformed pixel points on the edge where the coordinates of the deformed pixel point are located is greater than that of the rest of the edges in the horizontal and vertical directions, then it means that the edge where the i-th pixel point is located is more likely to be the edge generated by the reflection of the coating surface.

[0044] Step 4: According to the grayscale category in the neighborhood of the edge where each pixel is located, and the difference between the average grayscale level in the neighborhood of each pixel and the average grayscale level of the coating grayscale image where it is located, the reflection coverage authenticity of each pixel is obtained, and the reflection interference authenticity of each pixel is obtained by combining the highlight reflection coefficient and reflection reflection coefficient of each pixel, and then the preset initial weight of each pixel in the coating grayscale image at the current moment is corrected.

[0045] For old residential buildings, due to the backwardness of science and technology and design concepts in the past, there will be a large number of self-used and self-built equipment on the roofs of old residential buildings, such as water tanks, solar water heaters and large air-conditioning outdoor units of high-rise residents. Since these facilities and equipment are closely related to the lives of residents, they cannot be dismantled or moved when cleaning and repairing the roof during the roof waterproofing construction process. Therefore, during the spraying process, these facilities and equipment will form shadows on the roof under the influence of light. The shape of these shadows is mainly presented in the image as relatively straight edges, with few branches and other edges with a high degree of similarity, which is similar to the edges formed by reflections.

[0046] However, even though shadows may cover the coating, since spray painting operations mostly occur during the day, the rooftop lighting is relatively bright, resulting in a low overall dimness of the shadows. This obstruction of the coating's features in the image and, consequently, the degree of interference with the identification of these features is low. Therefore, during the denoising process using the NLM algorithm, these shadow-obscured pixels do not actually significantly interfere with the algorithm's performance. However, due to the high degree of similarity between the edge features formed by shadows and reflections, the aforementioned steps alone can easily misidentify shadow areas as reflections, leading to a reduction in the weights of these pixels. This results in a loss of analysis of these coating features, impacting the final image quality, and therefore requires further analysis.

[0047] Specifically, reflections created by light not only reveal the shape of objects on the coating surface, but also their corresponding colors. Therefore, the areas around the edges of the reflections experience significant color changes. This is reflected in the grayscale image of the coating, where the grayscale of pixels at or around certain edges changes significantly or forms localized clusters. (Under normal circumstances, the coating is a solid color, but if reflections occur, the reflected building appears in multiple colors, and the pixels at the edges of the reflections in the grayscale image can be divided into multiple grayscale regions.) Regarding shadows, the areas covered by the shadows experience only a slight decrease in brightness compared to the unobstructed areas of the coating.

[0048] To address the above characteristics, a v×v window centered on a single pixel is used as the neighborhood window for that pixel (v is an odd number, v=5 in this example, but can be adjusted by the implementer based on actual accuracy requirements). All pixels within the neighborhood window corresponding to all pixels on a single edge are considered the neighboring pixels of that edge. All neighboring pixels on a single edge are used as input to a region growing algorithm. The growing condition is set to begin when the grayscale difference between the seed and target points is less than Q, and the stopping condition is set to stop when the grayscale values ​​of the seed and target points are greater than or equal to Q. The neighboring pixel area of ​​a single edge is divided into multiple growing regions (Q is set to 15 in this example, but can be adjusted by the implementer based on actual accuracy requirements). The pixels contained in each growing region represent a specific color. Region growing algorithms are well-known techniques, and the detailed process is not detailed here. Furthermore, because the majority of pixels in the grayscale image of the coating captured during normal spraying are pixels of the coating's native color, the average grayscale value of the coating grayscale image can be used as the grayscale value closest to the coating's native grayscale.

[0049] As a preferred embodiment, the reflection coverage authenticity of each pixel point is obtained based on the grayscale category within the neighborhood range of the edge where each pixel point is located, and the difference between the average grayscale level in the neighborhood of each pixel point and the average grayscale level of the grayscale image of the coating where it is located, which is used to characterize the possibility that the characteristics of the area where each pixel point is located are the characteristics generated by reflection rather than the characteristics generated by shadow coverage.

[0050] In this embodiment, the reflection coverage authenticity of the i-th pixel is recorded as , its specific expression is: Where, is the reflection coverage authenticity of the i-th pixel, is the total number of growth areas of neighboring pixels on the edge of the i-th pixel, is the grayscale mean of all pixels in the neighborhood window of the i-th pixel, is the grayscale mean of the grayscale image of the coating where the i-th pixel is located. It should be noted that when the i-th pixel is not on the edge, is the gray value variance of all pixels in the neighborhood window of the i-th pixel.

[0051] The meaning of this relationship is: when the total number of growth areas formed by the neighboring pixels on the edge of the i-th pixel point is larger, or the grayscale value variance in the neighborhood window is larger, and the difference between the grayscale value of the pixel point in the neighborhood window of the pixel point and the overall grayscale value of the coating itself is more obvious, it means that the location of the pixel point is more likely to be a change caused by reflection rather than shadow coverage. In this case, the weight of the pixel point should be reduced when using the NLM algorithm for denoising to avoid distortion of the coating features.

[0052] As a preferred embodiment, the reflection interference authenticity of each pixel point is obtained based on the reflection coverage authenticity, highlight reflection coefficient and reflection reflection coefficient of each pixel point, which is used to characterize the intensity of the reflection interference suffered by each pixel point. The flowchart for obtaining the reflection interference authenticity of each pixel point is as follows: Figure 2 shown.

[0053] In this embodiment, the reflection interference authenticity of the i-th pixel is recorded as , its specific expression is: Where, is the reflection interference authenticity of the i-th pixel, is the highlight reflection coefficient of the i-th pixel, is the reflection coefficient of the i-th pixel, is the reflection coverage authenticity of the i-th pixel.

[0054] The meaning of this relationship is that when the i-th pixel point in the coating image is more likely to be the pixel point where the highlight reflection or reflective shadow is located on the coating surface, and the more likely it is to be covered by the reflective shadow, the stronger the reflective interference to the pixel point is. When using the NLM algorithm for denoising, the less the original coating features that can be represented by the pixel point, the lower the weight should be given.

[0055] Furthermore, the preset initial weight of each pixel in the current coating grayscale image is corrected, and the calculation formula of the corrected weight of each pixel is: Where, is the corrected weight of the i-th pixel, To preset the initial weight, is the reflection interference authenticity of the i-th pixel, norm() is the normalization function, and in this embodiment, the tanh function is used for normalization. It can be obtained by calculation using the NLM algorithm. The specific process is a well-known technology and will not be described in detail in this application.

[0056] At this point, the corrected weight of each pixel in the current coating grayscale image can be obtained.

[0057] Furthermore, the original coating grayscale image and the corrected weight of each pixel in the image are used as inputs of the NLM algorithm, and the denoised coating grayscale image is output. The NLM algorithm is a well-known technology and the specific process will not be described in detail.

[0058] After acquiring the denoised coating grayscale image data, the spray control platform of the automated spray equipment connects to a pre-built database of coating defect images and uses an image matching algorithm. The algorithm takes the denoised coating grayscale image and the coating defect image data as input and outputs a matching score. When the matching score is greater than or equal to a pre-set threshold (set to 0.95 in this solution, but adjustable based on actual accuracy requirements), it indicates a defect in the current coating grayscale image. The system then outputs the defect category based on the labeled defect type in the matched coating defect image data. Image matching algorithms are well-known technologies, and the detailed process is omitted here.

[0059] Furthermore, when the spray control platform identifies the type of defect in the coating, it sends a spray control signal. After the automated spray equipment receives the control signal, it sends a control signal to the pneumatic system, hydraulic system, etc. of the spray equipment, and finally controls and adjusts the relevant spray parameters.

[0060] For example, if cracking is identified as a defect, it could be due to insufficient spray pressure during the spraying process, resulting in poor paint atomization, coarse spray particles, and an inability to form a uniform film, leading to cracking. At this point, the spray control platform issues a pressure increase control signal. Upon receiving the control signal, the automated spray equipment further adjusts the air compressor output or the regulating valve opening to increase the spray pressure.

[0061] At this point, a waterproof coating spraying control method that improves construction adaptability has been completed.

[0062] Based on the same inventive concept as the above method, an embodiment of the present application also provides a waterproof coating spraying control system for improving construction adaptability, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any one of the above-mentioned waterproof coating spraying control methods for improving construction adaptability.

[0063] It should be noted that the order in which the embodiments of the present application are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential sequence shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0064] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

[0065] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A waterproof coating spraying control method for improving construction adaptability, characterized in that: The method comprises the following steps: Real-time acquisition of grayscale images of the waterproof coating during spraying; Comparing the grayscale value of each pixel with a preset segmentation threshold to obtain highlighted pixels; clustering all highlighted pixels in the current coating grayscale image, and obtaining neighboring clusters of each cluster based on the distance between each cluster and other clusters; obtaining the highlight reflectivity coefficient of each pixel based on the number of pixels in the cluster where each pixel is located, the average level of distances between the cluster and the remaining clusters, and the difference between the average grayscale value of the cluster where the pixel is located and the average grayscale value of its neighboring clusters; Obtain the edges and deformed pixels in the current coating grayscale image, and obtain the reflection coefficient of each pixel based on the similarity between the edge where each pixel is located and other edges, the number of endpoints on the edge, and the number of deformed pixels in the horizontal and vertical directions of each deformed pixel in the edge; According to the grayscale category in the neighborhood of the edge where each pixel is located, and the difference between the average grayscale level in the neighborhood of each pixel and the average grayscale level of the coating grayscale image where it is located, the reflection coverage authenticity of each pixel is obtained, and the reflection interference authenticity of each pixel is obtained by combining the highlight reflection coefficient and reflection reflection coefficient of each pixel, and then the preset initial weight of each pixel in the coating grayscale image at the current moment is corrected.

2. A waterproof coating spraying control method for improving construction adaptability according to claim 1, characterized in that: The highlighted pixel points refer to pixel points whose grayscale values ​​are greater than or equal to a preset segmentation threshold.

3. A waterproof coating spraying control method for improving construction adaptability according to claim 1, characterized in that: The process of obtaining the neighboring clusters of each cluster is as follows: calculating the Euclidean distance between the center points of a single cluster and all other clusters, and taking the clusters with Euclidean distances less than a preset neighboring threshold as the neighboring clusters of the cluster.

4. A waterproof coating spraying control method for improving construction adaptability according to claim 1, characterized in that: The calculation formula of the highlight reflection coefficient of each pixel is: Where, is the highlight reflection coefficient of the i-th pixel, is the total number of pixels contained in the cluster where the i-th pixel is located, is the mean of the Euclidean distances between the cluster where the i-th pixel is located and the center points of the remaining clusters, is the grayscale mean of the pixel in the cluster where the i-th pixel is located, is the grayscale mean of the pixels in the jth neighboring cluster of the cluster where the i-th pixel is located, and J is the total number of neighboring clusters of the cluster where the i-th pixel is located.

5. A waterproof coating spraying control method for improving construction adaptability according to claim 1, characterized in that: The process of acquiring the deformed pixel points is as follows: acquiring the curvature of each pixel point in each edge of the coating grayscale image at the current moment, and taking the pixel points whose curvature is greater than or equal to the preset deformation threshold as the deformed pixel points of the coating grayscale image at the current moment.

6. A waterproof coating spraying control method for improving construction adaptability according to claim 1, characterized in that: The calculation formula of the reflection coefficient of each pixel is: Where, is the reflection coefficient of the i-th pixel, is the mean of the cosine similarity between the edge where the i-th pixel is located and all other edges, is the total number of end edge points on the edge where the i-th pixel is located, is the total number of deformed pixels in the horizontal and vertical directions of the x-th deformed pixel in the edge where the i-th pixel is located. is the total number of deformed pixels in the edge where the i-th pixel is located; wherein, the end edge point on the edge refers to a pixel point that has only two edge pixels in the eight neighborhoods of the edge.

7. A waterproof coating spraying control method for improving construction adaptability according to claim 1, characterized in that: The calculation formula for the reflection coverage authenticity of each pixel is: Where, is the reflection coverage authenticity of the i-th pixel, is the total number of growth areas of neighboring pixels on the edge of the i-th pixel, is the grayscale mean of all pixels in the neighborhood window of the i-th pixel, is the grayscale mean of the grayscale image of the coating where the i-th pixel is located; wherein the neighborhood window of the i-th pixel refers to a window with the i-th pixel as the center and a preset distance as the radius; the neighboring pixels of the edge refer to all pixels within the neighborhood window of all pixels in the edge; the growth area of ​​the neighboring pixels of the edge refers to a plurality of growth areas into which the neighboring pixel areas of each edge are divided using a region growing algorithm.

8. A waterproof coating spraying control method for improving construction adaptability according to claim 1, characterized in that: The reflection interference authenticity of each pixel point refers to the forward fusion result of the reflection coverage authenticity, highlight reflection coefficient and reflection reflection coefficient of each pixel point.

9. A waterproof coating spraying control method for improving construction adaptability according to claim 1, characterized in that: The preset initial weight of each pixel in the current coating grayscale image is corrected, and the calculation formula of the corrected weight of each pixel is: Where, is the corrected weight of the i-th pixel, To preset the initial weight, is the reflection interference authenticity of the i-th pixel, and norm() is the normalization function.

10. A waterproof coating spraying control system for improving construction adaptability, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of a waterproof coating spraying control method for improving construction adaptability as described in any one of claims 1-9 are implemented.

Images