Wall decoration detection method and system

Abstract

The invention provides a wall decoration detection method and system, and relates to the technical field of building detection, and the method comprises the steps: obtaining wall decoration data information based on a laser level meter and a camera device, the data information comprising height, level and image information; the data information is sent to a wall flatness model for flatness calculation, and flatness error data is obtained. Edge calculation is carried out based on the flatness error data and the decoration data information, crack data are obtained, and the crack data comprise position and form information. And analyzing the data information through the data analysis model to obtain smoothness and color data. And finally, the flatness error data, the crack data, the smoothness and the color data are sent to a qualification analysis model for analysis, and a detection result of wall decoration is obtained. Through analysis and detection of various data, comprehensive evaluation of the wall decoration quality is realized, whether the wall is qualified or not can be accurately judged, and the method has the characteristics of high efficiency, accuracy and high automation degree.

Description

Technical Field

[0001] The present invention relates to the technical field of building detection, and in particular to a wall decoration detection method and system. Background Art

[0002] With the rapid development of the construction industry, the quality of wall decoration has become a focus of attention for users and construction parties. Existing wall decoration detection methods mainly rely on manual visual inspection and simple tool measurement, which cannot comprehensively and accurately evaluate quality indicators such as wall flatness, cracks and smoothness. This traditional method is not only inefficient, but also easily affected by human factors, resulting in inaccurate detection results. In addition, the existing technology lacks systematic analysis and processing of data in wall decoration quality detection, making it difficult to fully grasp the complex wall conditions, and thus unable to effectively discover potential quality risks, which brings difficulties to subsequent decoration and maintenance.

[0003] Therefore, there is an urgent need for a new wall decoration detection method that can comprehensively analyze indicators such as wall flatness, cracks, smoothness and color through multiple data acquisition methods to provide comprehensive and accurate detection results. Summary of the invention

[0004] The purpose of the present invention is to provide a wall decoration detection method and system to improve the above problems. In order to achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0005] In a first aspect, the present application provides a wall decoration detection method, comprising:

[0006] Acquire wall decoration data information based on a laser level and a camera device, wherein the wall decoration data information includes height data information, level data information and image information of all walls;

[0007] Sending the wall decoration data information to a wall flatness model for flatness calculation to obtain wall flatness error data;

[0008] Perform edge computing based on the flatness error data of the wall and the wall decoration data information to obtain crack data of the wall, wherein the crack data of the wall includes position data and shape data of the wall crack;

[0009] Sending the wall decoration data information to a data analysis model for data analysis to obtain smoothness data and color data of the wall;

[0010] The flatness error data, crack data, smoothness data and color data of the wall are sent to a qualified analysis model for data analysis to obtain a detection result of the wall decoration.

[0011] In a second aspect, the present application also provides a wall decoration detection system, comprising:

[0012] An acquisition unit, used for acquiring wall decoration data information based on a laser level and a camera device, wherein the wall decoration data information includes height data information, level data information and image information of all walls;

[0013] A first processing unit is used to send the wall decoration data information to a wall flatness model to perform flatness calculation to obtain wall flatness error data;

[0014] A second processing unit is used to perform edge computing based on the flatness error data of the wall and the wall decoration data information to obtain crack data of the wall, wherein the crack data of the wall includes position data and shape data of the wall crack;

[0015] A first analysis unit is used to send the wall decoration data information to a data analysis model for data analysis to obtain smoothness data and color data of the wall;

[0016] The second analysis unit is used to send the flatness error data, crack data, smoothness data and color data of the wall surface to the qualified analysis model for data analysis to obtain the detection result of the wall surface decoration.

[0017] The beneficial effects of the present invention are:

[0018] The present invention combines a laser level with a camera to comprehensively obtain wall decoration data information, including height data, level data and image information. By sending these data to a flatness model, a data analysis model and a qualified analysis model, a systematic flatness calculation, smoothness and color analysis and crack data extraction are performed to obtain a comprehensive detection result of the wall decoration. This method not only improves the detection efficiency and accuracy, but also overcomes the deficiencies of the prior art in data processing and analysis, realizes a comprehensive evaluation of the wall state, and provides important technical support for improving the quality of wall decoration.

[0019] Other features and advantages of the present invention will be set forth in the following description, and in part will become apparent from the description, or may be understood by practicing embodiments of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0021] Figure 1 It is a schematic diagram of the process of the wall decoration detection method described in an embodiment of the present invention;

[0022] Figure 2 It is a schematic diagram of the structure of the wall decoration detection system described in an embodiment of the present invention.

[0023] In the figure: 701, acquisition unit; 702, first processing unit; 703, second processing unit; 704, first analysis unit; 705, second analysis unit; 7021, first processing subunit; 7022, second processing subunit; 7023, third processing subunit; 7024, fourth processing subunit; 7025, first calculation subunit; 7031, fifth processing subunit; 7032, sixth processing subunit; 7033, seventh processing subunit; 7034, eighth processing subunit; 7035, eighth processing subunit The second computing subunit; 7036, the third computing subunit; 7041, the ninth processing subunit; 7042, the fourth computing subunit; 7043, the fifth computing subunit; 7044, the first analyzing subunit; 70441, the second analyzing subunit; 70442, the sixth computing subunit; 70443, the seventh computing subunit; 7051, the third analyzing subunit; 7052, the fourth analyzing subunit; 7053, the fifth analyzing subunit; 7054, the sixth analyzing subunit; 7055, the seventh analyzing subunit. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0025] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings. At the same time, in the description of the present invention, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0026] Embodiment 1:

[0027] This embodiment provides a wall decoration detection method.

[0028] See also Figure 1 , the figure shows that the method includes step S1, step S2, step S3, step S4 and step S5.

[0029] Step S1, obtaining wall decoration data information based on a laser level and a camera device, wherein the wall decoration data information includes height data information, horizontal data information and image information of all walls;

[0030] It is understandable that the laser level can accurately measure the height data of the wall. It usually uses the principle of light beam emission and reception, combined with a high-precision measurement algorithm to ensure that the height difference of the wall is captured over a large range. At the same time, it can avoid the human errors that may be introduced by traditional means in level detection and ensure the reliability of the data. The camera equipment is responsible for collecting the image information of the wall. Through high-resolution camera technology, it can capture the texture, color, cracks and other features of the wall in detail, thus laying a good foundation for subsequent image processing and analysis.

[0031] The data information obtained in this step, including height data information, horizontal data information and image information, provides a multi-dimensional perspective, enabling the subsequent analysis model to be comprehensively evaluated. Specifically, the height and horizontal data of the wall will be used to establish a preliminary flatness model, while the image information provides raw data for crack detection and color analysis. This comprehensive acquisition of data not only improves the comprehensiveness and accuracy of the detection, but also effectively identifies possible quality risks. In addition, in terms of technical effects, this step transforms wall detection from traditional qualitative analysis to quantitative analysis, greatly improving the scientificity and objectivity of the detection, and providing data support for subsequent qualified analysis.

[0032] Step S2, sending the wall decoration data information to the wall flatness model for flatness calculation to obtain wall flatness error data;

[0033] It can be understood that this step makes the assessment of wall flatness more scientific and objective through precise numerical calculation and model optimization, greatly improves the monitoring ability of wall decoration quality, and lays the foundation for subsequent quality analysis. In this step, step S2 includes step S21, step S22, step S23, step S24 and step S25.

[0034] Step S21, dividing each wall into at least two grids based on the orthogonal grid method;

[0035] It can be understood that the orthogonal grid method first divides the wall into multiple rectangular or square grid areas based on the overall size and shape of the wall. This division method is simple and easy to implement, so that each grid can effectively represent a part of the wall, thereby more clearly reflecting the height change and horizontal information of the wall. In practical applications, according to the actual measurement data of the wall, the size and number of grids can be flexibly adjusted to adapt to the complexity of different wall structures and the sparsity of data. In the division process, the starting point and end point of the grid are first determined, and then a coordinate system is established on the wall with uniform spacing. This coordinate system will help to accurately mark the subsequent height and horizontal data. The coordinate information of each grid will provide basic data for flatness calculation to ensure the accuracy of subsequent interpolation and error compensation processes. Through scientific grid division, the subsequent analysis is more comprehensive and the reliability of the overall detection is enhanced.

[0036] Step S22, marking each grid based on the height data information and horizontal data information of all the walls, and interpolating the height data information and horizontal data information of each grid point between adjacent grid points based on a piecewise linear interpolation method, wherein the height data and horizontal data of the intermediate interpolation point are generated according to the intermediate distance between two grid points, and an initial flatness model is constructed, wherein the initial flatness model includes continuous height data distribution data and horizontal data distribution data;

[0037] It can be understood that the linearization processing of the piecewise linear interpolation method solves the problems caused by the discreteness of the measured data, so that a continuous height and level distribution is formed over a large area of ​​the wall. This process not only helps to eliminate jumps and discontinuities between data, but also can more realistically reflect the actual state of the wall. Through efficient interpolation methods, subsequent error compensation and model fitting are more accurate, and at the same time, the overall state of the wall can be truly reflected, providing a reliable basis for subsequent flatness calculations. This not only improves the accuracy of the analysis, but also provides more comprehensive data support for the detection process, ensuring the reliability and effectiveness of the final detection results.

[0038] Step S23, performing error compensation on the initial flatness model based on a preset Gaussian process regression kernel function to obtain an error-compensated initial flatness model;

[0039] It can be understood that the Gaussian kernel function is often used to capture the smoothness of the data, and its characteristics enable it to adapt to changes in wall height and horizontal data. In addition, appropriate hyperparameters need to be set to optimize the fitting effect of the model. Subsequently, the initial flatness model is processed using the preset Gaussian process regression kernel function to calculate the error and corresponding correction value of each point in the model. This process can not only compensate for the measurement error, but also smooth the data, eliminate fluctuations caused by local extreme values ​​or noise, and make the performance of the model more stable. Among them, the error compensation method based on Gaussian process regression demonstrates its flexibility and advantages in processing complex data sets, making the results of the entire wall detection process more reliable and usable. In this step, the preset Gaussian process regression kernel function is as follows:

[0040]

[0041] Among them, k(x i , x j ) represents the input point x i and x j The covariance between represents the variance of the function value, l represents the length scale, ||x i -x j || 2 Represents the input point x i and x j The square of the Euclidean distance between them.

[0042] It can be understood that after determining the covariance through the Gaussian kernel function, the covariance matrix can be constructed through the covariance, and then the Bayesian theorem is used to give a new input point and calculate the mean of its posterior distribution to obtain the error and achieve error compensation.

[0043] Step S24, performing data transformation on each grid point in the initial flatness model after error compensation based on a spline surface fitting method to obtain a continuous flatness fitting surface;

[0044] It can be understood that the flatness fitting surface generated in this step can provide a reliable basis for subsequent analysis. Among them, the smoothness characteristic of spline fitting can more realistically reflect the actual state of the wall surface compared to the simple interpolation method.

[0045] The formula for spline surface fitting is as follows:

[0046]

[0047]

[0048] Where S(x, y) represents the height value of the generated spline surface at point (x, y); N i，p(u) represents the B-spline basis function, M j，q (υ) represents the corresponding spline basis function in the y direction, P ij represents a known point on the spline surface, E represents the total error, m represents the number of spline segments, n represents the total number of sample points, z i Indicates the actual height value of the sample point, S(x i ,y i ) represents the height predicted by the model.

[0049] Step S25: performing second-order derivative calculation on the continuous flatness fitting surface based on the Laplace operator to obtain flatness error data of each grid area of ​​the surface.

[0050] It is understandable that the Laplace operator provides local information for each grid point, showing the rate of change of the point compared to its neighborhood. If the Laplace value of a certain area is positive, it means that the point is a local minimum (concave), while a negative value means it is a local maximum (convex). This is crucial for the flatness detection of the wall, because excessive convexities or concavities may directly affect the quality and aesthetics of the decoration.

[0051] By calculating the flatness error data of each grid area, subsequent analysis can more accurately identify specific problems on the wall, such as obvious unevenness or local deformation, and provide a basis for subsequent repair and decoration. This calculation step not only improves the accuracy of wall detection, but also lays the foundation for the visualization and intelligent analysis of subsequent data, ensuring that the efficiency and effect of wall detection are significantly improved in actual applications.

[0052] The formula for calculating the second-order derivative is as follows:

[0053]

[0054] Among them, Z represents the height value of the wall flatness fitting surface at any point, X represents any horizontal coordinate Y on the flatness fitting surface, and Y represents any vertical coordinate on the flatness fitting surface. Represents the result of applying the Laplace operator to the flatness fitting surface.

[0055] Step S3, performing edge calculation based on the flatness error data of the wall and the wall decoration data information to obtain crack data of the wall, wherein the crack data of the wall includes position data and shape data of the wall crack;

[0056] It can be understood that this step improves the detection rate of cracks, and also makes the detection results more visual, which is convenient for subsequent analysis and processing, and provides a reliable basis for the subsequent extraction of crack position data and morphological data. In this way, the state of wall cracks can be quickly located and evaluated in real-time monitoring and analysis, thereby providing data support for subsequent repair and maintenance. In this step, step S3 includes step S31, step S32, step S33, step S34, step S35 and step S36.

[0057] Step S31, using the Canny edge detection algorithm to process the image information of the wall, and calculate the gradient of each pixel in the image information of the wall;

[0058] It is understandable that the Canny algorithm performs Gaussian smoothing on the original wall image in order to remove random noise from the image. By selecting a suitable Gaussian kernel, the algorithm effectively reduces the interference of noise on subsequent edge detection results. Next, the algorithm calculates the gradient of each pixel, usually using the Sobel operator, which is a commonly used edge detection operator that calculates the derivative of the image in the horizontal and vertical directions to obtain the gradient strength and direction of each pixel. The gradient strength reflects the brightness change of the point and other points in its neighborhood, while the gradient direction indicates the direction of the edge.

[0059] In this process, in the gradient image generated by the Canny algorithm, areas with higher intensity are marked as edge candidates, while areas with lower intensity are considered possible noise. On this basis, the next step will further apply threshold processing and edge connection to determine which points constitute the actual edge. Therefore, in terms of technical effect, by calculating the gradient of each pixel point, not only can the crack features in the wall image be efficiently extracted, but also reliable basic data can be provided for subsequent edge tracking and morphological analysis. The successful implementation of this process will lay a solid foundation for the subsequent acquisition of crack location and morphological data, making the overall wall detection process more accurate and efficient.

[0060] Step S32: distinguishing the gradient of each pixel point based on a preset threshold value to obtain strong edges and weak edges of the image information of the wall, and connecting the strong edges and weak edges of the image information of the wall by a hysteresis edge method to obtain continuous wall crack edge image information;

[0061] It can be understood that the system sets a high threshold and a low threshold. The gradient intensity of each pixel is compared with these two thresholds. If the gradient intensity of a pixel is greater than the high threshold, the pixel is considered a strong edge and is marked as a possible location of a crack. On the contrary, if the gradient intensity of the pixel is lower than the low threshold, it is considered noise and ignored. Pixels between the two thresholds are marked as weak edges, which require further analysis to determine whether to include them in the final edge result. After that, the system will start with the strong edge and perform a local search to check the adjacent weak edge pixels. If the gradient intensity of the adjacent weak edge pixel is greater than the low threshold, it is considered to be part of the connection. This method ensures that even relatively weak edges can be included in the final crack edge image information as long as they are connected to the strong edge. This process not only improves the robustness of edge detection, but also ensures the continuity and integrity of the final result, significantly reducing the risk of edge loss due to noise or small intensity changes.

[0062] Step S33, performing enhancement processing on the crack edge image information of the continuous wall surface based on morphological opening and closing operations to obtain enhanced crack edge image information;

[0063] It can be understood that the opening operation is a morphological operation, which uses a structural element to corrode the image and then perform an expansion operation. In the crack edge image, the application of the opening operation can effectively remove small noise points and minor interferences, and retain larger crack features. Specifically, by sliding a structural element (such as a circle or rectangle) on the image, the opening operation will first shrink the edges of all objects in the image, remove noise points smaller than the structural element, and then expand the image back to its original size to restore the larger crack edges. In this way, the main technical effect of the opening operation is to eliminate noise and smooth edges while retaining the main crack information. The closing operation is the inverse operation of the opening operation, which first dilates the image and then erodes it. By applying the closing operation on the crack edge image, the gaps or small breaks between the cracks can be effectively filled.

[0064] After the combination of opening and closing operations, the enhanced crack edge image information is clear and coherent, providing strong support for subsequent crack data analysis. This enhancement processing not only improves the visual effect of the image, but also enhances the algorithm's ability to extract crack features, thereby ensuring high accuracy and reliability of wall decoration detection.

[0065] Step S34: Mapping the pixel points of the crack from the image space to the parameter space based on Hough transform, and detecting the crack features with linear or arc shape using an accumulator;

[0066] It can be understood that the process of mapping the crack pixel points from the image space to the parameter space based on Hough transform in this step can be expressed by the following formula:

[0067]

[0068] Among them, P(r, θ) represents the accumulator value of a specific parameter combination, I(Q) represents the value of the current pixel, and H(x Q ,y Q , r, θ) represents the edge pixel (x Q ,y Q ) is mapped to the mapping function in the parameter space (r, θ), E represents the set of edge pixels, and Q represents the index of the edge pixel currently being processed.

[0069] Step S35, segmenting the crack area containing the crack features based on the watershed algorithm, and calculating the width and length of each crack;

[0070] It can be understood that this step maps the intensity value of each pixel in the image to its corresponding height by constructing a height map. The high-intensity area will form a "peak" in the height map, while the low-intensity area will form a "valley". Then, by simulating the process of water flow, it will flow from the peak to the valley, and the area segmentation will be determined according to the inflow of water. When the water flows meet, a watershed line will be formed at the meeting point, thereby separating different crack areas.

[0071] The formula for calculating the width and length of each crack is as follows;

[0072]

[0073]

[0074] Among them, L f represents the total length of the crack, W f represents the average width of the crack, M represents the total number of points on the crack boundary, (u k , v k ) represents the coordinates of the kth crack boundary point, g l Represents the width measurement at the lth point.

[0075] Step S36: Determine the position data of the wall cracks based on the flatness error data of the wall and the width and length of each crack.

[0076] It can be understood that this step maps the morphological information of the cracks into the three-dimensional coordinate system of the wall, thereby obtaining the precise location data of the cracks, which not only improves the accuracy of detection, but also provides effective data support for subsequent repair plans. This method is particularly suitable for dealing with cracks that may occur in complex wall structures, laying the foundation for the development of crack repair work.

[0077] Step S4: sending the wall decoration data information to a data analysis model for data analysis to obtain smoothness data and color data of the wall;

[0078] It can be understood that this step can obtain specific data on the smoothness and color of the wall surface, providing detailed basic information for subsequent quality inspection. This processing method based on the data analysis model can effectively quantify the appearance characteristics of the wall surface, which not only improves the accuracy of the analysis, but also makes the results have strong practical guidance, which is convenient for the implementation of subsequent processing and rectification measures. In this step, step S4 includes step S41, step S42, step S43 and step S44.

[0079] Step S41, sending the height data information and the horizontal data information of the wall to a three-dimensional rectangular coordinate system for processing to obtain a three-dimensional model of the wall;

[0080] It can be understood that this step can convert the two-dimensional wall data into a three-dimensional geometric model through three-dimensional rectangular coordinate system processing. This model provides a true geometric expression of the wall surface, and provides accurate basic data for subsequent wall surface characteristic analysis, such as smoothness and crack detection. At the same time, it also improves the system's ability to identify tiny defects on the wall surface, making the detection results more reliable and accurate.

[0081] Step S42, calculating the average value of adjacent grid points of each grid point in the three-dimensional rectangular coordinate system based on the local average method, and calculating the smoothness index of the grid point based on a preset smoothness index calculation formula;

[0082] It can be understood that the local average method in this step can smooth the three-dimensional space grid point data and eliminate the interference of noise, thereby improving the accuracy of smoothness calculation. Through the calculation of the smoothness index formula, the system can quantify the smoothness of the wall and identify local subtle defects. This method not only improves the detection accuracy of smoothness, but also provides a scientific basis for the subsequent determination of whether the wall is qualified as a whole. Among them, the preset smoothness index calculation formula is as follows:

[0083]

[0084] Among them, S represents the smoothness index of the current grid point, P i Indicates the height value of the current grid point, Q jrepresents the height value of the adjacent grid points, k represents the number of adjacent grid points, |P i -Q j |Indicates the absolute value of the height difference between the current grid point and the adjacent grid point.

[0085] Step S43, performing weighted average calculation based on the smoothness indexes of all grid points, and performing standard deviation calculation based on the calculated global smoothness value, and using the calculated standard deviation as the smoothness error data of the wall surface;

[0086] It can be understood that this step evaluates the smoothness error data of the wall surface by weighted averaging the smoothness index of all grid points and further calculating the standard deviation. This process can not only comprehensively reflect the overall smoothness of the entire wall surface, but also quantify the degree of deviation of the local area from the overall smoothness, thereby more accurately reflecting the uniformity of the wall surface smoothness.

[0087] Step S44: Perform color analysis based on the image information of the wall to obtain color data information of the wall.

[0088] It can be understood that this step can effectively identify the color quality of the wall and find possible color differences or unevenness through this color analysis method. For example, if the main color of a certain area is significantly different from the overall wall color, it may indicate uneven painting or material problems in the area. In this step, step S44 includes step S441, step S442 and step S443.

[0089] Step S441, performing edge detection on the image information of the wall to obtain the wall edge of the image information of the wall, and converting the wall image outside the wall edge into black;

[0090] It can be understood that this step can clearly identify the edge features of the wall through edge detection, making the subsequent crack analysis and wall quality assessment more accurate. This process can effectively remove irrelevant background information and focus on the main features of the wall structure, thereby providing high-quality basic data for subsequent analysis.

[0091] Step S442, clustering the three components of R, G, and B of all pixels within the edge of the wall image, and averaging the center points of all the obtained clusters to obtain the average value of the center points of all the clusters, and using the average value as the color data of the image information of the wall;

[0092] It can be understood that this step can effectively extract the color features of the wall and simplify the color representation of the wall by clustering the pixels within the edge. The clustering results will help to more accurately describe the color characteristics of the wall in practical applications, making color analysis and subsequent decoration quality assessment more scientific and systematic. This process not only reduces the data dimension, but also eliminates color interference caused by lighting changes or image noise, and enhances the stability and repeatability of color data.

[0093] Step S443, dividing the color data of the image information of the wall into regions according to the grid points in the three-dimensional rectangular coordinate system, and performing mean calculation on the color data corresponding to the divided regions to obtain the color data information of each grid point.

[0094] It can be understood that this step can effectively combine the wall color characteristics with the spatial structure by performing regional division and mean calculation on the color data of the wall image information. This method not only reduces the complexity of the data, but also improves the understanding of the overall visual effect of the wall. In practical applications, the color data information of each grid point obtained can provide important reference for subsequent decoration design, quality control and aesthetic evaluation.

[0095] Step S5: sending the flatness error data, crack data, smoothness data and color data of the wall surface to a qualified analysis model for data analysis to obtain a detection result of the wall surface decoration.

[0096] It is understandable that this step can comprehensively evaluate the quality of wall decoration by integrating and analyzing data from multiple dimensions. This comprehensive analysis not only improves the accuracy of the detection, but also makes the analysis process more transparent. In practical applications, the test results can provide a scientific basis for construction management, guide subsequent rectification measures, and ensure that the decoration quality meets the standard requirements. In this step, step S5 includes step S51, step S52, step S53, step S54 and step S55.

[0097] Step S51: Performing a hierarchical analysis on the flatness error data, crack data, smoothness data and color data of the wall surface based on a hierarchical analysis method to determine the weight of each type of data;

[0098] It can be understood that this step can effectively quantify the relative importance of various factors in the wall decoration quality assessment by assigning weights to different data types through the hierarchical analysis method. The advantage of this method is its systematicity and flexibility, which makes the assessment process more objective and scientific. In practical applications, this weight analysis helps to optimize the qualified analysis model, so that the model can more accurately reflect the actual situation of wall decoration when processing complex data, thereby improving the reliability and effectiveness of the overall detection.

[0099] Step S52: performing correlation analysis on all types of data and preset qualified thresholds to obtain correlation values ​​of each type of data;

[0100] It is understandable that by correlating each data type with the preset qualified threshold, the quality level of wall decoration can be quantitatively evaluated. The advantage of this method is that it can systematically integrate multiple data dimensions to form a comprehensive assessment of the wall decoration status. At the same time, correlation analysis can identify which indicators fail to meet the qualified standards, thereby providing a specific basis for subsequent improvement and repair measures. In practical applications, this process helps to promptly discover potential problems in wall decoration, ensure construction quality, and reduce subsequent maintenance costs.

[0101] Step S53: Based on the preset minimum discrimination principle, the weight of each type of data and its corresponding correlation value are summarized to obtain a summary weight value corresponding to each type of data;

[0102] It can be understood that the summary formula in this step is as follows

[0103]

[0104] Among them, Q a represents the redistribution weight of the a-th data type, U a represents the initial weight of the a-th data type, T a Indicates the association value of the a-th data type, Represents the weighted sum of the weights and relevance values ​​of all data types.

[0105] Step S54: score each type of data according to a preset level threshold, and multiply the scores obtained by the scoring by the corresponding summary weight values ​​to obtain a weighted score for each type of data;

[0106] It can be understood that this step combines the performance of each data type with its relative importance through weighted calculation, so that the final weighted score can accurately reflect the overall quality of the wall decoration. This method can effectively reduce the improper influence of a specific data type on the final result during the evaluation process, ensuring the comprehensiveness and objectivity of the test results.

[0107] Step S55: Based on the weighted score of each type of data, determine whether the data of this type on the wall is qualified, and obtain a determination result.

[0108] It can be understood that by comparing the weighted scores with the preset thresholds, we can clearly determine whether the wall surface is qualified in each quality indicator. This method ensures the objectivity and reliability of quality assessment, so that the weighted scores of various data can effectively reflect the overall quality status of the wall surface. The output judgment results of this step will provide a direct basis for subsequent qualified analysis and necessary rectification measures, and further improve the quality control and management level of wall surface decoration.

[0109] Embodiment 2:

[0110] like Figure 2 As shown, this embodiment provides a wall decoration detection system, see Figure 2 The system includes an acquisition unit 701 , a first processing unit 702 , a second processing unit 703 , a first analysis unit 704 and a second analysis unit 705 .

[0111] An acquisition unit 701 is used to acquire wall decoration data information based on a laser level and a camera device, wherein the wall decoration data information includes height data information, level data information and image information of all walls;

[0112] The first processing unit 702 is used to send the wall decoration data information to the wall flatness model to perform flatness calculation to obtain wall flatness error data;

[0113] The first processing unit 702 includes a first processing subunit 7021 , a second processing subunit 7022 , a third processing subunit 7023 , a fourth processing subunit 7024 and a first calculation subunit 7025 .

[0114] The first processing subunit 7021 is used to divide each wall surface into at least two grids based on the orthogonal grid method;

[0115] The second processing subunit 7022 is used to mark each grid based on the height data information and the horizontal data information of all the wall surfaces, and interpolate the height data information and the horizontal data information of each grid point between adjacent grid points based on the piecewise linear interpolation method, wherein the height data and the horizontal data of the intermediate interpolation point are generated according to the intermediate distance between two grid points, and an initial flatness model is constructed, wherein the initial flatness model includes continuous height data distribution data and horizontal data distribution data;

[0116] The third processing subunit 7023 is used to perform error compensation on the initial flatness model based on a preset Gaussian process regression kernel function to obtain an initial flatness model after error compensation;

[0117] The fourth processing subunit 7024 is used to perform data conversion on each grid point in the initial flatness model after error compensation based on a spline surface fitting method to obtain a continuous flatness fitting surface;

[0118] The first calculation subunit 7025 is used to calculate the second-order derivative of the continuous flatness fitting surface based on the Laplace operator to obtain flatness error data of each grid area of ​​the surface.

[0119] The second processing unit 703 is used to perform edge calculation based on the flatness error data of the wall and the wall decoration data information to obtain crack data of the wall, where the crack data of the wall includes position data and shape data of the wall crack;

[0120] The second processing unit 703 includes a fifth processing subunit 7031 , a sixth processing subunit 7032 , a seventh processing subunit 7033 , an eighth processing subunit 7034 , a second calculation subunit 7035 and a third calculation subunit 7036 .

[0121] The fifth processing subunit 7031 is used to process the image information of the wall using the Canny edge detection algorithm to calculate the gradient of each pixel in the image information of the wall;

[0122] The sixth processing subunit 7032 is used to distinguish the gradient of each pixel point based on a preset threshold value, obtain the strong edge and the weak edge of the image information of the wall, and connect the strong edge and the weak edge of the image information of the wall by a lag edge method, so as to obtain continuous wall crack edge image information;

[0123] The seventh processing subunit 7033 is used to enhance the crack edge image information of the continuous wall surface based on morphological opening and closing operations to obtain enhanced crack edge image information;

[0124] The eighth processing subunit 7034 is used to map the pixel points of the crack from the image space to the parameter space based on Hough transform, and detect the crack features with linear or arc shape by using the accumulator;

[0125] The second calculation subunit 7035 is used to segment the crack area containing the crack features based on the watershed algorithm and calculate the width and length of each crack;

[0126] The third calculation subunit 7036 is used to determine the position data of the wall cracks based on the flatness error data of the wall and the width and length of each crack.

[0127] The first analysis unit 704 is used to send the wall decoration data information to the data analysis model for data analysis to obtain the smoothness data and color data of the wall;

[0128] The first analyzing unit 704 includes a ninth processing subunit 7041 , a fourth calculating subunit 7042 , a fifth calculating subunit 7043 and a first analyzing subunit 7044 .

[0129] The ninth processing subunit 7041 is used to send the height data information and the horizontal data information of the wall to the three-dimensional rectangular coordinate system for processing to obtain a three-dimensional model of the wall;

[0130] The fourth calculation subunit 7042 is used to calculate the average value of adjacent grid points of each grid point in the three-dimensional rectangular coordinate system based on the local average method, and calculate the smoothness index of the grid point based on a preset smoothness index calculation formula;

[0131] The fifth calculation subunit 7043 is used to calculate the weighted average value based on the smoothness indexes of all grid points, and calculate the standard deviation based on the calculated global smoothness value, and use the calculated standard deviation as the smoothness error data of the wall surface;

[0132] The first analysis subunit 7044 is used to perform color analysis based on the image information of the wall to obtain color data information of the wall.

[0133] The first analyzing subunit 7044 includes a second analyzing subunit 70441 , a sixth calculating subunit 70442 and a seventh calculating subunit 70443 .

[0134] The second analysis subunit 70441 is used to perform edge detection on the image information of the wall to obtain the wall edge of the image information of the wall, and convert the wall image outside the wall edge into black;

[0135] The sixth calculation subunit 70442 is used to perform clustering processing on the three components of R, G, and B of all pixels within the edge of the wall image, and perform average calculation on the center points of all the obtained clusters to obtain the average value of the center points of all the clusters, and use the average value as the color data of the image information of the wall;

[0136] The seventh calculation subunit 70443 is used to divide the color data of the image information of the wall into regions according to the grid points in the three-dimensional rectangular coordinate system, and perform mean calculation on the color data corresponding to the divided regions to obtain the color data information of each grid point.

[0137] The second analysis unit 705 is used to send the flatness error data, crack data, smoothness data and color data of the wall surface to the qualified analysis model for data analysis to obtain the detection result of the wall surface decoration.

[0138] The second analysis unit 705 includes a third analysis subunit 7051 , a fourth analysis subunit 7052 , a fifth analysis subunit 7053 , a sixth analysis subunit 7054 and a seventh analysis subunit 7055 .

[0139] The third analysis subunit 7051 is used to perform hierarchical analysis on the flatness error data, crack data, smoothness data and color data of the wall surface based on the hierarchical analysis method to determine the weight of each type of data;

[0140] The fourth analysis subunit 7052 is used to obtain a correlation value of each type of data based on correlation analysis of all types of data with a preset qualified threshold value;

[0141] The fifth analysis subunit 7053 is used to summarize the weight of each type of data and its corresponding association value based on a preset minimum discrimination principle to obtain a summary weight value corresponding to each type of data;

[0142] The sixth analysis subunit 7054 is used to score each type of data according to a preset level threshold, and multiply the scores obtained by the scoring by the corresponding summary weight values ​​to obtain a weighted score for each type of data;

[0143] The seventh analysis subunit 7055 is used to judge whether the data of each type of the wall surface is qualified based on the weighted score of the data of that type, and obtain a judgment result.

[0144] It should be noted that, regarding the system in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0145] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0146] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A wall decoration detection method, characterized in that: include: Acquire wall decoration data information based on a laser level and a camera device, wherein the wall decoration data information includes height data information, level data information and image information of all walls; Sending the wall decoration data information to a wall flatness model for flatness calculation to obtain wall flatness error data; Perform edge computing based on the flatness error data of the wall and the wall decoration data information to obtain crack data of the wall, wherein the crack data of the wall includes position data and shape data of the wall crack; Sending the wall decoration data information to a data analysis model for data analysis to obtain smoothness data and color data of the wall; The flatness error data, crack data, smoothness data and color data of the wall are sent to a qualified analysis model for data analysis to obtain a detection result of the wall decoration.

2. The wall decoration detection method according to claim 1 is characterized in that , sending the wall decoration data information to the wall flatness model for flatness calculation to obtain the wall flatness error data, including: Divide each wall into at least two grids based on an orthogonal grid method; Marking each grid based on the height data information and the horizontal data information of all the walls, and interpolating the height data information and the horizontal data information of each grid point between adjacent grid points based on a piecewise linear interpolation method, wherein the height data and the horizontal data of the intermediate interpolation point are generated according to the intermediate distance between two grid points, and an initial flatness model is constructed, wherein the initial flatness model includes continuous height data distribution data and horizontal data distribution data; Performing error compensation on the initial flatness model based on a preset Gaussian process regression kernel function to obtain an error-compensated initial flatness model; Based on the spline surface fitting method, data transformation is performed on each grid point in the initial flatness model after error compensation to obtain a continuous flatness fitting surface; The second-order derivative of the continuous flatness fitting surface is calculated based on the Laplace operator to obtain flatness error data of each grid area of ​​the surface.

3. The wall decoration detection method according to claim 1 is characterized in that , performing edge computing based on the flatness error data of the wall and the wall decoration data information, including: The Canny edge detection algorithm is used to process the image information of the wall, and the gradient of each pixel in the image information of the wall is calculated; The gradient of each pixel point is distinguished based on a preset threshold value to obtain strong edges and weak edges of the image information of the wall, and the strong edges and weak edges of the image information of the wall are connected by a hysteresis edge method to obtain continuous wall crack edge image information; Based on morphological opening and closing operations, the crack edge image information of the continuous wall surface is enhanced to obtain enhanced crack edge image information; Based on Hough transform, the pixel points of the crack are mapped from the image space to the parameter space, and the accumulator is used to detect the crack features with linear or arc shape; The crack area containing crack features is segmented based on the watershed algorithm, and the width and length of each crack are calculated; The position data of the wall cracks are determined based on the flatness error data of the wall and the width and length of each crack.

4. The wall decoration detection method according to claim 1, characterized in that , sending the wall decoration data information to the data analysis model for data analysis to obtain the smoothness data and color data of the wall, including: The height data information and the horizontal data information of the wall are sent to a three-dimensional rectangular coordinate system for processing to obtain a three-dimensional model of the wall; The average value of adjacent grid points of each grid point in the three-dimensional rectangular coordinate system is calculated based on the local average method, and the smoothness index of the grid point is calculated based on a preset smoothness index calculation formula; A weighted average is calculated based on the smoothness index of all grid points, and a standard deviation is calculated based on the calculated global smoothness value, and the calculated standard deviation is used as the smoothness error data of the wall surface; Color analysis is performed based on the image information of the wall to obtain color data information of the wall.

5. The wall decoration detection method according to claim 4 is characterized in that ,Based on the image information of the wall, color analysis is performed to obtain the color data information of the wall, including: Perform edge detection on the image information of the wall to obtain the wall edge of the image information of the wall, and convert the wall image outside the wall edge into black; The three components of R, G, and B of all pixels within the edge of the wall image are clustered separately, and the average value of the center points of all the obtained clusters is calculated to obtain the average value of the center points of all the clusters, and the average value is used as the color data of the image information of the wall; The color data of the image information of the wall is divided into regions according to the grid points in the three-dimensional rectangular coordinate system, and the color data corresponding to the divided regions are averaged to obtain the color data information of each grid point.

6. A wall decoration detection system, characterized in that: include: An acquisition unit, used for acquiring wall decoration data information based on a laser level and a camera device, wherein the wall decoration data information includes height data information, level data information and image information of all walls; A first processing unit is used to send the wall decoration data information to a wall flatness model to perform flatness calculation to obtain wall flatness error data; A second processing unit is used to perform edge computing based on the flatness error data of the wall and the wall decoration data information to obtain crack data of the wall, wherein the crack data of the wall includes position data and shape data of the wall crack; A first analysis unit is used to send the wall decoration data information to a data analysis model for data analysis to obtain smoothness data and color data of the wall; The second analysis unit is used to send the flatness error data, crack data, smoothness data and color data of the wall surface to the qualified analysis model for data analysis to obtain the detection result of the wall surface decoration.

7. The wall decoration detection system according to claim 6, characterized in that: The first processing unit comprises: A first processing subunit, configured to divide each wall surface into at least two grids based on an orthogonal grid method; The second processing subunit is used to mark each grid based on the height data information and the horizontal data information of all the wall surfaces, and interpolate the height data information and the horizontal data information of each grid point between adjacent grid points based on a piecewise linear interpolation method, wherein the height data and the horizontal data of the intermediate interpolation point are generated according to the intermediate distance between two grid points, and an initial flatness model is constructed, wherein the initial flatness model includes continuous height data distribution data and horizontal data distribution data; A third processing subunit is used to perform error compensation on the initial flatness model based on a preset Gaussian process regression kernel function to obtain an initial flatness model after error compensation; The fourth processing subunit is used to perform data conversion on each grid point in the initial flatness model after error compensation based on a spline surface fitting method to obtain a continuous flatness fitting surface; The first calculation subunit is used to calculate the second-order derivative of the continuous flatness fitting surface based on the Laplace operator to obtain flatness error data of each grid area of ​​the surface.

8. The wall decoration detection system according to claim 6, characterized in that: The second processing unit comprises: A fifth processing subunit is used to process the image information of the wall using a Canny edge detection algorithm to calculate the gradient of each pixel in the image information of the wall; a sixth processing subunit, configured to distinguish the gradient of each pixel point based on a preset threshold value, obtain strong edges and weak edges of the image information of the wall, and connect the strong edges and weak edges of the image information of the wall by a hysteresis edge method, thereby obtaining continuous crack edge image information of the wall; A seventh processing subunit is used to enhance the crack edge image information of the continuous wall surface based on morphological opening and closing operations to obtain enhanced crack edge image information; An eighth processing subunit, used for mapping the pixel points of the crack from the image space to the parameter space based on Hough transform, and detecting the crack features with linear or arc shape by using an accumulator; The second calculation subunit is used to segment the crack area containing the crack features based on the watershed algorithm and calculate the width and length of each crack; The third calculation subunit is used to determine the position data of the wall cracks based on the flatness error data of the wall and the width and length of each crack.

9. The wall decoration detection system according to claim 6, characterized in that: The first analysis unit comprises: A ninth processing subunit, configured to send the height data information and the horizontal data information of the wall to a three-dimensional rectangular coordinate system for processing to obtain a three-dimensional model of the wall; A fourth calculation subunit is used to calculate the average value of adjacent grid points of each grid point in the three-dimensional rectangular coordinate system based on a local average method, and calculate the smoothness index of the grid point based on a preset smoothness index calculation formula; A fifth calculation subunit is used to calculate a weighted average value based on the smoothness indexes of all grid points, and to calculate a standard deviation based on the calculated global smoothness value, and to use the calculated standard deviation as the smoothness error data of the wall surface; The first analysis subunit is used to perform color analysis based on the image information of the wall to obtain color data information of the wall.

10. The wall decoration detection system according to claim 9, characterized in that: The first analysis subunit comprises: The second analysis subunit is used to perform edge detection on the image information of the wall to obtain the wall edge of the image information of the wall, and convert the wall image outside the wall edge into black; a sixth calculation subunit, configured to perform clustering processing on the three components of R, G, and B of all pixels within the edge of the image of the wall, and perform average calculation on the center points of all the obtained clusters to obtain the average value of the center points of all the clusters, and use the average value as the color data of the image information of the wall; The seventh calculation subunit is used to divide the color data of the image information of the wall into regions according to the grid points in the three-dimensional rectangular coordinate system, and perform mean value calculation on the color data corresponding to the divided regions to obtain the color data information of each grid point.

Images