Insulation board adhesive area ratio detection method, device, terminal and storage medium

Abstract

The present application discloses a method, device, terminal and storage medium for detecting the adhesive area ratio of an insulation board, wherein the method comprises: obtaining a first image of an insulation board and a second image of a wall corresponding to the insulation board, and obtaining the total area of ​​the insulation board according to the first image; preprocessing and registering and fusing the first image and the second image respectively, and obtaining a first comprehensive image of the insulation board and a second comprehensive image of the wall; detecting the first comprehensive image and the second comprehensive image respectively according to an edge detection algorithm and obtaining a detection result, and obtaining a first bonding area of ​​the first comprehensive image and a second bonding area of ​​the second comprehensive image according to the detection result; identifying and counting the pixel points in the first bonding area and the second bonding area, and obtaining the actual bonding area of ​​the insulation board, and calculating the bonding area ratio of the insulation board in combination with the total area. The method of the present application can automatically and accurately calculate the actual bonding area, and improve the efficiency and accuracy of the calculation results.

Description

Technical Field

[0001] The present application relates to the technical field of building material detection, and in particular to a method, device, terminal and storage medium for detecting the adhesive area ratio of an insulation board. Background Art

[0002] At present, in the construction industry, the detection of the bonding area ratio of insulation boards is an important link to ensure the thermal insulation performance and structural safety of walls. Traditional detection methods usually rely on manual visual inspection or use a 10mm spacing grid plate to count the number of squares in the bonding area one by one. This process is not only time-consuming and laborious, but also has a large number of subjective judgment factors, resulting in limited detection accuracy.

[0003] Existing solutions mainly include manual grid measurement and radar detection. Although the manual grid measurement method is intuitive, it is inefficient in large-scale detection, while the radar detection method is mainly used for the detection of hidden projects and has poor detection effect on non-metallic materials such as insulation boards.

[0004] Therefore, the existing technology has obvious deficiencies in detecting the adhesive area ratio of the insulation board, especially the traditional manual detection method, which leads to low work efficiency, high cost, and is easily affected by human factors, resulting in unstable detection results. In addition, although the radar detection method is effective under certain conditions, its wide application is limited to the detection of non-metallic materials, so it is also limited in detection in this field.

[0005] In summary, how to achieve more efficient and accurate insulation board adhesive area ratio detection is an urgent problem to be solved in this field. Summary of the invention

[0006] In order to improve the efficiency and accuracy of insulation board bonding area detection, the present application provides an insulation board bonding area ratio detection method, device, terminal and storage medium. By acquiring multi-perspective images of the insulation board and the corresponding wall, the area where the adhesive exists is identified, and by processing the identification results, the actual bonding area is accurately calculated, thereby improving the efficiency and accuracy of the calculation results.

[0007] In the first aspect, the present application provides a method for detecting the adhesive area ratio of an insulation board, which adopts the following technical solution:

[0008] A method for detecting the adhesive area ratio of an insulation board comprises the following steps:

[0009] Acquire a first image of an insulation board and a second image of a wall corresponding to the insulation board, and obtain a total area of ​​the insulation board according to the first image, wherein the first image and the second image include adhesives carried on the insulation board and the corresponding wall;

[0010] Preprocessing and registering and fusing the first image to obtain a first comprehensive image of the insulation board, and preprocessing and registering and fusing the second image to obtain a second comprehensive image of the wall corresponding to the insulation board;

[0011] According to a preset edge detection algorithm, the first integrated image and the second integrated image carrying the adhesive are respectively detected and detection results are obtained, and according to the detection results, a first bonding area of ​​the first integrated image and a second bonding area of ​​the second integrated image are obtained;

[0012] The pixel points in the first bonding area and the second bonding area are identified and counted to obtain the actual bonding area of ​​the insulation board, and the bonding area ratio of the insulation board is calculated in combination with the total area.

[0013] By adopting the above technical solution, starting from multi-view image recognition of insulation boards and walls, through edge segmentation, 3D reconstruction, registration and fusion, until the final calculation of the actual adhesive area, a complete and highly integrated processing flow is formed, ensuring that each step from the original image to the final result can be processed accurately and reliably, ensuring the efficiency and accuracy of the detection process; at the same time, through technologies such as image recognition and automated detection processes, the detection efficiency is greatly improved, and the labor cost and risk of human error are significantly reduced.

[0014] Preferably, the step of acquiring a first image of the insulation board and a second image of the wall corresponding to the insulation board, and obtaining the total area of ​​the insulation board according to the first image, specifically comprises the following steps:

[0015] Acquire a first image of the insulation board at a plurality of preset viewing angles, and a second image of a wall corresponding to the insulation board at a plurality of the preset viewing angles, wherein the wall corresponding to the insulation board is an area bonded by the adhesive before the insulation board is peeled off, and the total area of ​​the insulation board is the same as the area of ​​the corresponding wall;

[0016] Each of the first image and the second image includes part or all of the adhesive, and records the camera parameters at each preset viewing angle.

[0017] By adopting the above technical solution, collecting images from multiple perspectives is conducive to capturing the overall shape of the adhesive, providing a data basis for the subsequent identification of the adhesive region and calculation of the adhesive area, thereby improving the accuracy of the adhesive area calculation.

[0018] Preferably, the first image is preprocessed and registered and fused to obtain a first comprehensive image of the insulation board, and the second image is preprocessed and registered and fused to obtain a second comprehensive image of the wall corresponding to the insulation board, which specifically includes the following steps:

[0019] Preprocessing all of the first images to obtain a first processed image, and preprocessing all of the second images to obtain a second processed image, wherein the preprocessing includes grayscale conversion, denoising filtering, and adaptive local binarization;

[0020] According to a preset image registration algorithm, in combination with the camera parameters, the first processed images of multiple viewing angles are aligned, and according to a preset image fusion algorithm, the aligned first processed images are fused to obtain a first comprehensive image, wherein the first comprehensive image includes the complete adhesive in the insulation board;

[0021] According to the image registration algorithm and in combination with the camera parameters, the second processed images of multiple perspectives are aligned, and the aligned second processed images are fused according to the image fusion algorithm to obtain a second comprehensive image, wherein the second comprehensive image includes the complete adhesive in the wall corresponding to the insulation board.

[0022] By adopting the above technical solution, all images are registered to ensure that they are spatially consistent, and the registered images are fused to generate a high-resolution image containing the complete adhesive area, thereby improving the accuracy of adhesive area calculation.

[0023] Preferably, the detecting the first integrated image and the second integrated image carrying the adhesive respectively according to a preset edge detection algorithm and obtaining the detection results specifically comprises the following steps:

[0024] Detecting the first integrated image according to a preset edge detection algorithm, wherein the first integrated image includes image information of the adhesive on the insulation board, and obtaining a first bonding boundary of the insulation board;

[0025] Detecting the second integrated image according to the edge detection algorithm, wherein the second integrated image includes image information of the adhesive remaining on the wall corresponding to the insulation board, and obtaining a second bonding boundary of the wall corresponding to the insulation board;

[0026] The first bonding boundary and the second bonding boundary are combined to obtain a detection result.

[0027] By adopting the above technical solution, the edge detection algorithm is used to extract the boundaries of the comprehensive image of the insulation board and the wall, which provides an accurate data source for the subsequent segmentation of the bonding area and the non-bonding area.

[0028] Preferably, obtaining the first bonding area of ​​the first comprehensive image and the second bonding area of ​​the second comprehensive image according to the detection result specifically comprises the following steps:

[0029] Performing morphological operations on the first integrated image and the second integrated image respectively, wherein the morphological operations include:

[0030] Respectively identifying and acquiring the number of noise points and holes in the first integrated image and the second integrated image, when the number of noise points in the first integrated image is greater than the number of holes, performing a closing operation on the first integrated image and obtaining a first operation result, and when the number of noise points is not greater than the number of holes, performing an opening operation on the first integrated image and obtaining the first operation result;

[0031] When the number of noise points in the second integrated image is greater than the number of holes, a closing operation is performed on the second integrated image to obtain a second operation result; when the number of noise points is not greater than the number of holes, an opening operation is performed on the second integrated image to obtain the second operation result;

[0032] The insulation board is divided into a first bonding area and a first non-bonding area according to the first calculation result and the first bonding boundary, and the wall corresponding to the insulation board is divided into a second bonding area and a second non-bonding area according to the second calculation result and the second bonding boundary.

[0033] By adopting the above technical solution, when the number of noise points is greater than the number of holes, the closing operation helps to fill small holes or connect adjacent areas. When the number of noise points is not greater than the holes, the opening operation helps to remove small noise points or disconnect narrow connections. The morphological operation method is used to optimize the final detection results and improve the accuracy of region segmentation.

[0034] Preferably, identifying and counting the pixel points in the first bonding area and the second bonding area to obtain the actual bonding area of ​​the thermal insulation board specifically comprises the following steps:

[0035] Identify the first bonding area to obtain a first bonding image, identify the second bonding area to obtain a second bonding image, and count the number of all pixels in the first bonding image and the second bonding image;

[0036] Performing feature extraction on the first bonding image and the second bonding image respectively and obtaining feature extraction results, performing three-dimensional reconstruction on the adhesive on the insulation board according to the feature extraction results and obtaining a three-dimensional model of board surface adhesion, and performing three-dimensional reconstruction on the adhesive on the wall corresponding to the insulation board according to the feature extraction results and obtaining a three-dimensional model of wall surface adhesion;

[0037] Performing point depth judgment on the board surface adhesive three-dimensional model and the wall surface adhesive three-dimensional model, wherein the point depth judgment includes judging whether there are pixel points whose depth is less than a preset first depth threshold in the board surface adhesive three-dimensional model and the wall surface adhesive three-dimensional model;

[0038] If not, the point depth is judged to be passed, and the actual bonding area of ​​the thermal insulation board is calculated based on the pixel points in the first bonding image and the second bonding image.

[0039] By adopting the above technical solution, three-dimensional modeling of the adhesive on the insulation board and the wall is carried out to obtain the depth of each pixel in the model, which assists in judging whether the adhesive at each pixel can play a role in actual bonding, thereby calculating the actual adhesive area that can play a role, which is more applicable to actual applications; at the same time, the logic of point depth judgment is set, which can save computing resources and time when the judgment conditions are met, and provides a more direct and efficient method for calculating the area of ​​the actual adhesive area.

[0040] Preferably, if there are pixels whose depth is less than the first depth threshold in the board surface adhesive three-dimensional model and the wall surface adhesive three-dimensional model, the point depth judgment fails, and the following steps are included:

[0041] Recording the point coordinates of the pixel points that are less than the first depth threshold and marking the pixel points as adhesion test points;

[0042] The three-dimensional adhesive model of the panel surface and the three-dimensional adhesive model of the wall surface are registered and fused to obtain a three-dimensional adhesive model, and the depth of the adhesive test point corresponding to each point coordinate in the three-dimensional adhesive model is measured to obtain the actual adhesive depth of each adhesive test point;

[0043] Determining whether all the adhesive test points meet a preset adhesive area calculation standard, wherein the adhesive area calculation standard includes that when the actual adhesive depth of any of the adhesive test points is less than a preset second depth threshold, the adhesive test point does not meet the adhesive area calculation standard and is not counted in the actual adhesive area, otherwise it is counted in the actual adhesive area;

[0044] The number of all the adhesion test points that meet the adhesion area calculation standard is counted, combined with the number of the pixel points that are not less than the first depth threshold, to calculate the actual adhesion area.

[0045] By adopting the above technical solution, when the conditions for point depth judgment are not met, the two three-dimensional models are aligned and fused to obtain a three-dimensional model of the complete adhesive, so as to determine whether the points that do not meet the conditions for point depth judgment meet the adhesive area calculation standards, and whether the points can play a role in actual bonding, so as to calculate the actual adhesive area, so that the final adhesive area is more reliable and accurate, thereby improving the accuracy of the calculation.

[0046] In the second aspect, the present application provides a device for detecting the adhesive area ratio of an insulation board, which adopts the following technical solution:

[0047] A device for detecting the adhesive area ratio of an insulation board, comprising the following modules:

[0048] An original image acquisition module, used to acquire a first image of the insulation board and a second image of the wall corresponding to the insulation board, and obtain the total area of ​​the insulation board according to the first image, wherein the first image and the second image include the adhesive carried on the insulation board and the corresponding wall;

[0049] A comprehensive image generation module, used to perform preprocessing and registration fusion on the first image to obtain a first comprehensive image of the insulation board, and to perform preprocessing and registration fusion on the second image to obtain a second comprehensive image of the wall corresponding to the insulation board;

[0050] a bonding area division module, configured to detect the first integrated image and the second integrated image carrying the adhesive according to a preset edge detection algorithm and obtain detection results, and obtain a first bonding area of ​​the first integrated image and a second bonding area of ​​the second integrated image according to the detection results;

[0051] The adhesive area calculation module is used to identify and count the pixel points in the first adhesive area and the second adhesive area to obtain the actual adhesive area of ​​the insulation board, and calculate the adhesive area ratio of the insulation board in combination with the total area.

[0052] By adopting the above-mentioned technical means, a system for identifying and detecting the adhesive area based on the original images of the insulation board and the wall was built, which provided the necessary software technical support for the accurate identification of the adhesive area between the two and ensured the safety and reliability of the construction.

[0053] In a third aspect, the present application provides a smart terminal, which adopts the following technical solution:

[0054] A smart terminal includes a memory and a processor, wherein the memory stores at least one instruction, at least one program, code set or instruction set, and the at least one instruction, at least one program, code set or instruction set is loaded and executed by the processor to implement the insulation board adhesion area ratio detection method as described above.

[0055] In a fourth aspect, the present application provides a computer-readable storage medium, which adopts the following technical solution:

[0056] A computer-readable storage medium, wherein at least one instruction, at least one program, code set or instruction set is stored in the computer-readable storage medium, and the at least one instruction, at least one program, code set or instruction set is loaded and executed by a processor to implement the insulation board adhesive area ratio detection method as described above.

[0057] In summary, this application has at least the following beneficial effects:

[0058] (1) This application starts with multi-view image recognition of insulation boards and walls, and goes through multiple steps such as edge segmentation, three-dimensional reconstruction, registration and fusion, until the final calculation of the actual adhesive area, forming a complete and highly integrated processing flow, ensuring that each step from the original image to the final result can be processed accurately and reliably, ensuring the efficiency and accuracy of the detection process; at the same time, through technologies such as image recognition and automated detection processes, the detection efficiency is greatly improved, and the labor cost and risk of human error are significantly reduced.

[0059] (2) This application can generate a high-resolution image containing the complete bonding area by collecting images from multiple perspectives and performing registration and fusion. At the same time, the logic of edge processing algorithm and morphological algorithm is adopted to segment and process the image, making the image more conducive to subsequent recognition and calculation, thereby improving the efficiency and accuracy of actual bonding area calculation.

[0060] (3) This application performs three-dimensional modeling of the insulation board and the wall, and takes the depth of the adhesive into consideration when calculating the adhesive area. This can eliminate pixel points that do not have bonding ability, making the final calculation of the adhesive area more accurate. At the same time, the logic of point depth judgment and adhesive area calculation standards is combined to improve the efficiency of automatic calculation, save unnecessary computing resources and time, and improve the efficiency, accuracy and comprehensiveness of area calculation. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 It is a flow chart of a method for detecting the adhesive area ratio of an insulation board according to an embodiment of the present application;

[0062] Figure 2 It is a flow chart of image registration and fusion of a method for detecting adhesive area ratio of thermal insulation board according to an embodiment of the present application;

[0063] Figure 3 is a flow chart of step S4 of the method for detecting the adhesive area ratio of the thermal insulation board according to an embodiment of the present application;

[0064] Figure 4 This is one of the structural diagrams of the device for detecting the adhesive area ratio of the thermal insulation board according to the embodiment of the present application;

[0065] Figure 5 This is the second structural diagram of the insulation board adhesion area ratio detection device in the embodiment of the present application. DETAILED DESCRIPTION

[0066] The present application provides a method, device, terminal and storage medium for detecting the adhesive area ratio of an insulation board. In order to make the purpose, technical solution and advantages of the present application clearer, the implementation method of the present application will be further described in detail below.

[0067] The following is a further detailed description of an embodiment of a method for detecting the adhesive area ratio of an insulation board of the present application in conjunction with the drawings of the specification.

[0068] Before going into a detailed description, it should be noted that the method of this embodiment is based on the insulation boards and walls of the building surface. Before detection, the insulation boards of a certain area need to be peeled off from the wall to which they are bonded. Therefore, the wall and the insulation boards correspond to each other, and the detected areas of the insulation boards and walls are the same. However, due to various factors, a portion of adhesive remains on the insulation boards and walls after peeling. Many technical solutions ignore the remaining portion on the wall when calculating the adhesive area. This embodiment takes into account the synergistic effect of the two, and therefore the technical solution of this application is based on the above background.

[0069] A method for detecting the adhesive area ratio of an insulation board in the present application, such as Figure 1 As shown, the following steps are included:

[0070] S1. Obtain a first image of an insulation board and a second image of a wall corresponding to the insulation board, and obtain the total area of ​​the insulation board based on the first image. The first image and the second image include adhesives carried on the insulation board and the corresponding wall, specifically including the following steps.

[0071] S11, acquiring first images of the insulation board at multiple preset viewing angles.

[0072] In this embodiment, a high-precision camera and a fixed bracket are used to take high-definition photos of the peeled insulation board from multiple preset viewing angles to ensure that part or all of the bonding area is captured, thereby obtaining multiple original first images of the insulation board.

[0073] S12: Acquire a second image of the wall corresponding to the insulation board at a plurality of preset viewing angles.

[0074] A high-precision camera is also used to take pictures of the wall after the insulation board is peeled off to ensure that the remaining adhesive portion of the wall is captured, thereby obtaining multiple original second images of the wall.

[0075] The wall corresponding to the insulation board is the area bonded by the adhesive before the insulation board is peeled off, and the total area of ​​the insulation board is the same as the area of ​​the corresponding wall.

[0076] Each of the first image and the second image includes at least a portion of adhesive.

[0077] S13. Record the camera parameters at each preset viewing angle.

[0078] The camera parameters include internal parameters and external parameters. The internal parameters include focal length, principal point coordinates and distortion parameters. The external parameters are used for the position and orientation of the camera in the world coordinate system and are composed of a rotation matrix and a translation vector. These camera parameters will be used for subsequent image registration.

[0079] S2, preprocessing and registering the first image to obtain a first comprehensive image of the insulation board, and preprocessing and registering the second image to obtain a second comprehensive image of the wall corresponding to the insulation board, specifically including the following steps.

[0080] S21, preprocessing all first images to obtain first processed images, and preprocessing all second images to obtain second processed images, wherein the preprocessing includes grayscale conversion, denoising filtering, and adaptive local binarization.

[0081] The first image and the second image are respectively subjected to adaptive local binarization after grayscale conversion and denoising filtering, which can enhance the contrast between the first image and the second image and solve the problem of poor detection effect in complex lighting environments, thereby ensuring the stability and consistency of the detection results.

[0082] In another specific embodiment, image enhancement technology is referenced, and a histogram normalization method is used to perform secondary enhancement on the contrast of all first images and second images after adaptive local binarization, thereby improving the global contrast and obtaining a first processed image corresponding to the first image and a second processed image corresponding to the second image, thereby further enhancing the image quality and recognition, making the image details more prominent, and being more helpful for subsequent image segmentation.

[0083] S22. Align the first processed images of multiple perspectives according to a preset image registration algorithm and in combination with camera parameters.

[0084] The process of image registration in this embodiment is as follows:

[0085] S221: Extract significant feature points from the first processed images at different viewing angles, such as corner points, edges, and contours of the image. In this embodiment, significant feature points are set as corner points.

[0086] S222: Use a scale-invariant feature transform (SIFT) algorithm to find corresponding corner point pairs in the first processed images at different viewing angles.

[0087] S223, using the matched feature point pairs and camera parameters, calculating the spatial transformation relationship between the first processed images of different perspectives, specifically including finding the best rotation matrix and translation vector so that the corner point pairs are aligned in space, and performing distortion correction on the first processed image according to the distortion parameters of the camera to ensure accurate matching of the corner points.

[0088] S224. Finally, the calculated spatial transformation relationship is used to rotate, translate and scale the first processed image, and the first processed images at different viewing angles are registered so that they are aligned in space.

[0089] S23, fusing the aligned first processed images according to a preset image fusion algorithm to obtain a first comprehensive image, wherein the first comprehensive image contains the complete adhesive in the insulation board, specifically comprising the following steps.

[0090] The multiple registered first processed images are merged into one image through image fusion to integrate the information of each first processed image, so as to obtain the final first integrated image.

[0091] The steps of image fusion in this embodiment are:

[0092] All the first processed images are fused according to a preset fusion strategy. The fusion strategy in this embodiment is a weighted average method.

[0093] In order to simplify the description process, the fusion process is described using two first processed images obtained after the two first images (first image P1 and first image P2) are registered.

[0094] The first processed images after the two registrations are I1 and I2, and the first integrated image after fusion is F. The weighted average fusion formula is used to fuse the two images. The weighted average fusion formula is:

[0095] F(x,y)=α*I1(x1,y1)+(1-α)*I2(x2,y2);

[0096] Where (x, y) represents the pixel position in the first composite image, (x1, y1) represents the pixel position in the I1 image, (x2, y2) represents the pixel position in the I2 image, and α is a weight coefficient between 0 and 1, which is used to control the contribution of I1 and I2 to the fusion result.

[0097] In practical applications, the specific value of the weight coefficient α is dynamically adjusted according to the quality or importance of the first processed images I1 and I2.

[0098] like Figure 2 The figure shows the registration and fusion process of this embodiment. The first integrated image after fusion includes all the bonding areas in the insulation board.

[0099] S24: align the second processed images of multiple perspectives according to the image registration algorithm and in combination with the camera parameters. The process of this step is the same as the above step S22, and will not be repeated here.

[0100] S25, fuse the aligned second processed images according to the image fusion algorithm to obtain a second comprehensive image, which contains the complete adhesive in the wall corresponding to the insulation board. The process of this step is the same as the above step S23, and will not be repeated here.

[0101] The fused second composite image includes all bonding areas in the wall corresponding to the insulation board.

[0102] S3, detecting the first integrated image and the second integrated image carrying the adhesive respectively according to a preset edge detection algorithm and obtaining detection results, and obtaining a first bonding area of ​​the first integrated image and a second bonding area of ​​the second integrated image according to the detection results, specifically comprising the following steps.

[0103] S31. Detect the first integrated image according to a preset edge detection algorithm to obtain a first bonding boundary of the insulation board. Specifically, the first bonding boundary is determined by applying a Canny edge detection operator.

[0104] The first integrated image includes image information of the adhesive on the insulation board.

[0105] S32. Detect the second integrated image according to an edge detection algorithm to obtain a second bonding boundary of the wall corresponding to the insulation board. Specifically, the second bonding boundary is determined by applying a Canny edge detection operator.

[0106] The second comprehensive image includes image information of adhesive left on the wall corresponding to the insulation board.

[0107] S33, summarizing the first bonding boundary and the second bonding boundary to obtain a detection result.

[0108] S34, performing morphological operations on the first integrated image and the second integrated image respectively, the morphological operations comprising:

[0109] S341 , respectively identify and obtain the number of noise points and holes in the first integrated image and the second integrated image.

[0110] S342, when the number of noise points in the first integrated image is greater than the number of holes, performing a closing operation on the first integrated image and obtaining a first operation result;

[0111] When the number of noise points is not greater than the number of holes, an opening operation is performed on the first integrated image to obtain a first operation result.

[0112] S343, when the number of noise points in the second integrated image is greater than the number of holes, performing a closing operation on the second integrated image and obtaining a second operation result;

[0113] When the number of noise points is not greater than the number of holes, an opening operation is performed on the second integrated image to obtain a second operation result.

[0114] When the number of noise points is greater than the number of holes, a closing operation is used to fill small holes or connect adjacent areas; when the number of noise points is not greater than the number of holes, an opening operation is used to remove small noise points or disconnect narrow connections.

[0115] S35, dividing the insulation board into a first bonding area and a first non-bonding area according to the first calculation result and the first bonding boundary;

[0116] The wall corresponding to the insulation board is divided into a second bonding area and a second non-bonding area according to the second calculation result and the second bonding boundary.

[0117] The above morphological operation is beneficial to image segmentation, making the segmentation of the bonding area and the non-bonding area more accurate.

[0118] S4, identify and count the pixel points in the first bonding area and the second bonding area to obtain the actual bonding area of ​​the insulation board, and calculate the bonding area ratio of the insulation board based on the total area, such as Figure 3 As shown, the specific steps include the following steps.

[0119] S41 , identifying the first bonding area to obtain a first bonding image, identifying the second bonding area to obtain a second bonding image, and counting the number of all pixels in the first bonding image and the second bonding image.

[0120] S42 , performing feature extraction on the first bonding image and the second bonding image respectively and obtaining feature extraction results.

[0121] The feature extraction is the same as step S221. The feature extraction result may be a feature point, a feature line or a region. In this embodiment, the extracted feature is a feature point.

[0122] S43, performing three-dimensional reconstruction of the adhesive on the insulation board according to the feature extraction result and obtaining a three-dimensional model of the board surface adhesive;

[0123] Based on the feature extraction results, the adhesive on the wall corresponding to the insulation board is three-dimensionally reconstructed and a three-dimensional model of the wall adhesive is obtained.

[0124] Taking the three-dimensional model of the board surface adhesive as an example, in this embodiment, the structured light method is used to reconstruct the adhesive on the insulation board in three dimensions. It is necessary to calculate the distance between the camera optical center and the feature point through the principle of triangulation to obtain the depth information of the generated feature point.

[0125] Specifically, a laser projector is used to project a series of point-shaped structured lights onto the insulation board, and then a high-resolution camera is used to capture the points illuminated by the structured light. These points are the feature points. By calculating the positional relationship between these points in the camera coordinate system and the world coordinate system, the three-dimensional shape and depth information of the adhesive on the insulation board are obtained, thereby reconstructing its three-dimensional model.

[0126] The process of establishing the three-dimensional model of wall adhesive is the same as above.

[0127] S44, performing point depth judgment on the board surface adhesive 3D model and the wall surface adhesive 3D model, wherein the point depth judgment includes judging whether there are pixel points whose depth is less than a preset first depth threshold in the board surface adhesive 3D model and the wall surface adhesive 3D model.

[0128] The purpose of point depth judgment is to avoid unnecessary calculations and save computing resources. If the point depth judgment meets the standard, it means that every pixel in the 3D model can play a role in the actual bonding, and the board surface bonding 3D model and the wall bonding 3D model can obtain the actual bonding area without subsequent fusion; if the point depth judgment does not meet the standard, it means that the bonding depth of some pixels in the model is not enough, and it is impossible to judge whether it can play a role in the actual bonding, so it is necessary to combine the bonding depth of the same point of another model to judge whether the combined depth can play a role in the actual bonding.

[0129] For example, if the amount of adhesive used in some areas is insufficient, it may only leave adhesive marks on the insulation board, but fail to bond to the same point on the wall. Therefore, this point cannot play a bonding role and should not be included in the actual bonding area.

[0130] S441: If there are no pixels less than the first depth threshold in the board surface adhesive 3D model and the wall surface adhesive 3D model, the point depth judgment is passed, and the actual adhesive area of ​​the insulation board is calculated based on the pixels in the first adhesive image and the second adhesive image, which specifically includes the following steps:

[0131] S4411 , obtaining a union image of the first and second bonded images based on the first and second bonded images.

[0132] The union image includes all pixels in the first bonded image and the second bonded image, and an overlapping area of ​​the first bonded image and the second bonded image.

[0133] In this embodiment, the overlapping area is obtained by comparing the pixel values ​​of the first bonding image and the second bonding image at the same pixel point, specifically,

[0134] The first bonding image and the second bonding image are binarized. That is, the bonding area is set to a white pixel value, and the non-bonding area is set to a black pixel value.

[0135] If the pixel values ​​of the two images at the same pixel are both white, then the location belongs to the overlapping area.

[0136] S4412, respectively calculating the glue area of ​​the union image and the overlapping area.

[0137] The edge detection algorithm is used to extract the contour of the adhesive area. According to the extracted contour, the number of pixels inside the contour is calculated, and the adhesive area of ​​the union image is calculated; according to the number of white pixels, the adhesive area of ​​the overlapping area is calculated.

[0138] S4413. Subtract the adhesive area of ​​the overlapping area from the adhesive area of ​​the union image to obtain the actual adhesive area of ​​the insulation board.

[0139] S442: If there are pixels whose depth is less than the first depth threshold in the board surface adhesive 3D model and the wall surface adhesive 3D model, the point depth determination fails, including the following steps:

[0140] S4421. Record the point coordinates of the pixel points that are less than the first depth threshold and mark the pixel points as adhesion test points.

[0141] S4422. Align and fuse the board surface adhesive 3D model and the wall surface adhesive 3D model to obtain a 3D adhesive model.

[0142] Specifically, the steps include matching the extracted feature points, establishing a corresponding relationship between the first bonding area and the second bonding area based on the feature points, calculating a transformation matrix between the two three-dimensional point clouds, and aligning the two point clouds to the same coordinate system using the transformation matrix;

[0143] The two point clouds are finely registered using the iterative closest point algorithm (ICP) to ensure that they are spatially consistent. The two registered point clouds are fused to generate a three-dimensional adhesive model that includes the complete adhesive area. This three-dimensional adhesive model is a three-dimensional model that combines the current insulation board and all the adhesives on the wall.

[0144] S4423. Measure the depth of the adhesive test point corresponding to each point coordinate in the three-dimensional adhesive model and obtain the actual adhesive depth of each adhesive test point (if a spatial coordinate system is established for the three-dimensional adhesive model, the actual adhesive depth is the Z coordinate in the spatial coordinate system).

[0145] In a specific embodiment, the measured actual adhesive depth is displayed in a visual manner to generate a three-dimensional depth map, which can intuitively show the depth distribution and changes of the adhesive area.

[0146] S4424, judging whether all the adhesive test points meet the preset adhesive area calculation standard, the adhesive area calculation standard includes:

[0147] When the actual adhesive depth of any adhesive test point is less than the preset second depth threshold, it means that the actual adhesive depth of the point cannot play a role in the actual bonding, and the adhesive test point does not meet the adhesive area calculation standard and is not included in the actual adhesive area;

[0148] On the contrary, when the actual adhesion depth of the adhesion test point is not less than the second depth threshold, the adhesion test point meets the adhesion area calculation standard and is counted into the actual adhesion area.

[0149] The second depth threshold is a value obtained based on experience. In this embodiment, the second depth threshold is greater than the first depth threshold.

[0150] S4425: Count the number of all adhesive test points that meet the adhesive area calculation standard, and calculate the actual adhesive area in combination with the number of pixel points that are not less than the first depth threshold.

[0151] Specifically, the sum of the number of all adhesive test points that meet the adhesive area calculation standard and the number of pixel points that are not less than the first depth threshold is calculated, and the actual adhesive area is calculated based on the sum of the numbers.

[0152] S45. Calculate the adhesive area ratio of the insulation board based on the total area of ​​the insulation board.

[0153] The total area of ​​the insulation board is the area of ​​the current detection area;

[0154] Adhesive area ratio = actual adhesive area / total area.

[0155] The adhesive area ratio calculated by the above method is not an area ratio obtained only visually, but a more reliable area ratio calculated by taking into account the adhesive area that can actually work.

[0156] Based on the same inventive concept as above, the present application also discloses a device for detecting the adhesive area ratio of an insulation board, the structure of which is as follows: Figure 4 and Figure 5 As shown, the device includes the following modules:

[0157] An original image acquisition module, used to acquire a first image of the insulation board and a second image of the wall corresponding to the insulation board, and obtain the total area of ​​the insulation board according to the first image, wherein the first image and the second image include adhesives carried on the insulation board and the corresponding wall;

[0158] A comprehensive image generation module is used to perform preprocessing and registration fusion on the first image to obtain a first comprehensive image of the insulation board, and to perform preprocessing and registration fusion on the second image to obtain a second comprehensive image of the wall corresponding to the insulation board;

[0159] The bonding area division module is used to detect the first integrated image and the second integrated image carrying the adhesive according to a preset edge detection algorithm and obtain detection results, and obtain a first bonding area of ​​the first integrated image and a second bonding area of ​​the second integrated image according to the detection results;

[0160] The adhesive area calculation module is used to identify and count the pixel points in the first adhesive area and the second adhesive area to obtain the actual adhesive area of ​​the insulation board, and calculate the adhesive area ratio of the insulation board in combination with the total area.

[0161] In a specific implementation scheme, the original image acquisition module includes the following units:

[0162] The first original image acquisition unit is used to acquire a first image of the insulation board at multiple preset viewing angles, and a second image of the wall corresponding to the insulation board at multiple preset viewing angles. The wall corresponding to the insulation board is an area bonded with an adhesive before the insulation board is peeled off. The total area of ​​the insulation board is the same as the area of ​​the corresponding wall. Each of the first image and the second image includes part or all of the adhesive.

[0163] The second original image acquisition unit is used to record the camera parameters at each preset viewing angle.

[0164] In a specific implementation scheme, the comprehensive image generation module includes the following units:

[0165] The first comprehensive image generating unit is used to preprocess all the first images to obtain the first processed images, and preprocess all the second images to obtain the second processed images, wherein the preprocessing includes grayscale conversion, denoising filtering and adaptive local binarization.

[0166] The second comprehensive image generating unit is used to align the first processed images of multiple perspectives according to a preset image registration algorithm and in combination with camera parameters, and to fuse the aligned first processed images according to a preset image fusion algorithm to obtain a first comprehensive image, wherein the first comprehensive image contains the complete adhesive in the insulation board.

[0167] The third comprehensive image generating unit is used to align the second processed images of multiple perspectives according to the image registration algorithm and the camera parameters, and fuse the aligned second processed images according to the image fusion algorithm to obtain a second comprehensive image, wherein the second comprehensive image includes the complete adhesive in the wall corresponding to the insulation board.

[0168] In a specific implementation scheme, the bonding area division module includes the following units:

[0169] A first bonding area division unit is used to detect the first comprehensive image according to a preset edge detection algorithm, the first comprehensive image includes image information of the adhesive on the insulation board, and obtain a first bonding boundary of the insulation board;

[0170] Detecting the second integrated image according to an edge detection algorithm, the second integrated image includes image information of the adhesive left on the wall corresponding to the insulation board, and obtaining a second bonding boundary of the wall corresponding to the insulation board;

[0171] The first bonding boundary and the second bonding boundary are combined to obtain a detection result.

[0172] The second bonding area division unit is used to perform morphological operations on the first integrated image and the second integrated image respectively, and the morphological operations include:

[0173] Respectively identifying and acquiring the number of noise points and holes in the first integrated image and the second integrated image, when the number of noise points in the first integrated image is greater than the number of holes, performing a closing operation on the first integrated image and obtaining a first operation result, and when the number of noise points is not greater than the number of holes, performing an opening operation on the first integrated image and obtaining a first operation result;

[0174] When the number of noise points in the second composite image is greater than the number of holes, a closing operation is performed on the second composite image to obtain a second operation result; when the number of noise points is not greater than the number of holes, an opening operation is performed on the second composite image to obtain a second operation result.

[0175] The third bonding area division unit is used to divide the insulation board into a first bonding area and a first non-bonding area according to the first calculation result and the first bonding boundary, and to divide the wall corresponding to the insulation board into a second bonding area and a second non-bonding area according to the second calculation result and the second bonding boundary.

[0176] In a specific implementation scheme, the adhesive area calculation module includes the following units:

[0177] The first adhesive area calculation unit is used to identify the first adhesive area to obtain a first adhesive image, identify the second adhesive area to obtain a second adhesive image, and count the number of all pixels in the first adhesive image and the second adhesive image.

[0178] The second adhesive area calculation unit is used to perform feature extraction on the first bonding image and the second bonding image respectively and obtain feature extraction results, perform three-dimensional reconstruction of the adhesive on the insulation board based on the feature extraction results and obtain a three-dimensional model of board surface adhesion, and perform three-dimensional reconstruction of the adhesive on the wall corresponding to the insulation board based on the feature extraction results and obtain a three-dimensional model of wall adhesion.

[0179] The third adhesive area calculation unit is used to perform point depth judgment on the board surface adhesive three-dimensional model and the wall surface adhesive three-dimensional model, and the point depth judgment includes judging whether there are pixel points less than a preset first depth threshold in the board surface adhesive three-dimensional model and the wall surface adhesive three-dimensional model;

[0180] If it does not exist, the point depth is judged to be passed, and the actual bonding area of ​​the insulation board is calculated based on the pixel points in the first bonding image and the second bonding image.

[0181] The fourth adhesive area calculation unit is used to determine that the point depth judgment fails if there are pixels less than the first depth threshold in the board surface adhesive three-dimensional model and the wall surface adhesive three-dimensional model.

[0182] Record the point coordinates of the pixel points that are less than the first depth threshold and mark the pixel points as adhesive test points;

[0183] The three-dimensional adhesive model of the board surface and the three-dimensional adhesive model of the wall surface are registered and fused to obtain a three-dimensional adhesive model, and the depth of the adhesive test point corresponding to each point coordinate in the three-dimensional adhesive model is measured to obtain the actual adhesive depth of each adhesive test point;

[0184] Determine whether all the adhesive test points meet the preset adhesive area calculation standard. The adhesive area calculation standard includes that when the actual adhesive depth of any adhesive test point is less than the preset second depth threshold, the adhesive test point does not meet the adhesive area calculation standard and is not included in the actual adhesive area. Otherwise, it is included in the actual adhesive area.

[0185] The fifth adhesive area calculation unit is used to count the number of all adhesive test points that meet the adhesive area calculation standard, and calculate the actual adhesive area in combination with the number of pixel points that are not less than the first depth threshold.

[0186] From the above function introduction, it can be seen that an insulation board adhesive area ratio detection device in this application has built a set of adhesive area recognition and detection system based on the original images of the insulation board and the wall, which provides the necessary software technical support for the accurate identification of the adhesive area between the two, provides guarantee for the safety and reliability of construction, and promotes the sustained and rapid development of the economy and society.

[0187] Based on the same inventive concept mentioned above, an embodiment of the present application also discloses a computer-readable storage medium, which stores at least one instruction, at least one program, code set or instruction set. The at least one instruction, at least one program, code set or instruction set can be loaded and executed by a processor to implement the insulation board adhesion area ratio detection method provided in the above method embodiment.

[0188] Also based on the same inventive concept mentioned above, an embodiment of the present application also discloses a computer-readable storage medium, which stores at least one instruction, at least one program, code set or instruction set. The at least one instruction, at least one program, code set or instruction set is loaded and executed by a processor to implement the insulation board adhesion area ratio detection method as described above.

[0189] Those skilled in the art will appreciate that all or part of the steps to implement the above embodiments may be accomplished by hardware or by instructing related hardware through a program, and the program may be stored in the computer-readable storage medium, which may include, for example, various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0190] The above description is only an optional embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for detecting the adhesive area ratio of an insulation board, characterized in that: The steps include: Acquire a first image of the insulation board at a plurality of preset viewing angles, and a second image of a wall corresponding to the insulation board at a plurality of the preset viewing angles, wherein the wall corresponding to the insulation board is an area bonded by an adhesive before the insulation board is peeled off, and obtain a total area of ​​the insulation board according to the first image, wherein the first image and the second image include the adhesive carried on the insulation board and the corresponding wall; Preprocessing and registering and fusing the first image to obtain a first comprehensive image of the insulation board, and preprocessing and registering and fusing the second image to obtain a second comprehensive image of the wall corresponding to the insulation board; According to a preset edge detection algorithm, the first integrated image and the second integrated image carrying the adhesive are respectively detected and detection results are obtained, the first integrated image and the second integrated image are respectively subjected to morphological operations and operation results are obtained, and a first bonding area of ​​the first integrated image and a second bonding area of ​​the second integrated image are obtained according to the detection results and the operation results; The pixel points in the first bonding area and the second bonding area are identified and counted to obtain the actual bonding area of ​​the insulation board, and the bonding area ratio of the insulation board is calculated in combination with the total area.

2. The method for detecting the adhesive area ratio of the thermal insulation board according to claim 1, characterized in that: The total area of ​​the insulation board is the same as the area of ​​the corresponding wall; Each of the first image and the second image includes part or all of the adhesive, and records the camera parameters at each preset viewing angle.

3. The method for detecting the adhesive area ratio of the thermal insulation board according to claim 2, characterized in that: Preprocessing and registering the first image to obtain a first comprehensive image of the insulation board, and preprocessing and registering the second image to obtain a second comprehensive image of the wall corresponding to the insulation board, specifically includes the following steps: Preprocessing all of the first images to obtain a first processed image, and preprocessing all of the second images to obtain a second processed image, wherein the preprocessing includes grayscale conversion, denoising filtering, and adaptive local binarization; According to a preset image registration algorithm, in combination with the camera parameters, the first processed images of multiple viewing angles are aligned, and according to a preset image fusion algorithm, the aligned first processed images are fused to obtain a first comprehensive image, wherein the first comprehensive image includes the complete adhesive in the insulation board; According to the image registration algorithm and in combination with the camera parameters, the second processed images of multiple perspectives are aligned, and the aligned second processed images are fused according to the image fusion algorithm to obtain a second comprehensive image, wherein the second comprehensive image includes the complete adhesive in the wall corresponding to the insulation board.

4. The method for detecting the adhesive area ratio of the thermal insulation board according to claim 2, characterized in that: The detecting of the first integrated image and the second integrated image carrying the adhesive according to a preset edge detection algorithm and obtaining the detection results specifically comprises the following steps: Detecting the first integrated image according to a preset edge detection algorithm, wherein the first integrated image includes image information of the adhesive on the insulation board, and obtaining a first bonding boundary of the insulation board; Detecting the second integrated image according to the edge detection algorithm, wherein the second integrated image includes image information of the adhesive remaining on the wall corresponding to the insulation board, and obtaining a second bonding boundary of the wall corresponding to the insulation board; The first bonding boundary and the second bonding boundary are combined to obtain a detection result.

5. The method for detecting the adhesive area ratio of the thermal insulation board according to claim 4, characterized in that: Performing morphological operations on the first integrated image and the second integrated image respectively and obtaining operation results, and obtaining a first bonding area of ​​the first integrated image and a second bonding area of ​​the second integrated image according to the detection results and the operation results, specifically includes the following steps: The morphological operation includes respectively identifying and obtaining the number of noise points and holes in the first integrated image and the second integrated image, and when the number of noise points in the first integrated image is greater than the number of holes, performing a closing operation on the first integrated image and obtaining a first operation result, and when the number of noise points is not greater than the number of holes, performing an opening operation on the first integrated image and obtaining the first operation result; When the number of noise points in the second integrated image is greater than the number of holes, a closing operation is performed on the second integrated image to obtain a second operation result; when the number of noise points is not greater than the number of holes, an opening operation is performed on the second integrated image to obtain the second operation result; The insulation board is divided into a first bonding area and a first non-bonding area according to the first calculation result and the first bonding boundary, and the wall corresponding to the insulation board is divided into a second bonding area and a second non-bonding area according to the second calculation result and the second bonding boundary.

6. The method for detecting the adhesive area ratio of the thermal insulation board according to claim 2, characterized in that: Identifying and counting the pixel points in the first bonding area and the second bonding area to obtain the actual bonding area of ​​the thermal insulation board specifically includes the following steps: Identify the first bonding area to obtain a first bonding image, identify the second bonding area to obtain a second bonding image, and count the number of all pixels in the first bonding image and the second bonding image; Performing feature extraction on the first bonding image and the second bonding image respectively and obtaining feature extraction results, performing three-dimensional reconstruction on the adhesive on the insulation board according to the feature extraction results and obtaining a three-dimensional model of board surface adhesion, and performing three-dimensional reconstruction on the adhesive on the wall corresponding to the insulation board according to the feature extraction results and obtaining a three-dimensional model of wall surface adhesion; Performing point depth judgment on the board surface adhesive three-dimensional model and the wall surface adhesive three-dimensional model, wherein the point depth judgment includes judging whether there are pixel points whose depth is less than a preset first depth threshold in the board surface adhesive three-dimensional model and the wall surface adhesive three-dimensional model; If not, the point depth is judged to be passed, and the actual bonding area of ​​the thermal insulation board is calculated based on the pixel points in the first bonding image and the second bonding image.

7. The method for detecting the adhesive area ratio of the thermal insulation board according to claim 6, characterized in that: If there are pixel points whose depth is less than the first depth threshold in the board surface adhesive three-dimensional model and the wall surface adhesive three-dimensional model, the point depth judgment fails, and the method includes the following steps: Recording the point coordinates of the pixel points that are less than the first depth threshold and marking the pixel points as adhesion test points; The three-dimensional adhesive model of the panel surface and the three-dimensional adhesive model of the wall surface are registered and fused to obtain a three-dimensional adhesive model, and the depth of the adhesive test point corresponding to each point coordinate in the three-dimensional adhesive model is measured to obtain the actual adhesive depth of each adhesive test point; Determining whether all the adhesive test points meet a preset adhesive area calculation standard, wherein the adhesive area calculation standard includes that when the actual adhesive depth of any of the adhesive test points is less than a preset second depth threshold, the adhesive test point does not meet the adhesive area calculation standard and is not counted in the actual adhesive area, otherwise it is counted in the actual adhesive area; The number of all the adhesion test points that meet the adhesion area calculation standard is counted, combined with the number of the pixel points that are not less than the first depth threshold, to calculate the actual adhesion area.

8. A device for detecting the adhesive area ratio of an insulation board, characterized in that: Includes the following modules: an original image acquisition module, used to acquire first images of the insulation board at a plurality of preset viewing angles, and second images of the wall corresponding to the insulation board at a plurality of the preset viewing angles, wherein the wall corresponding to the insulation board is an area bonded by an adhesive before the insulation board is peeled off, and the total area of ​​the insulation board is obtained according to the first image, and the first image and the second image include the adhesive carried on the insulation board and the corresponding wall; A comprehensive image generation module, used for preprocessing and registering the first image to obtain a first comprehensive image of the insulation board, and preprocessing the second image to obtain a second comprehensive image of the wall corresponding to the insulation board; a bonding area division module, configured to detect the first integrated image and the second integrated image carrying the adhesive according to a preset edge detection algorithm and obtain detection results, perform morphological operations on the first integrated image and the second integrated image and obtain operation results, and obtain a first bonding area of ​​the first integrated image and a second bonding area of ​​the second integrated image according to the detection results and the operation results; The adhesive area calculation module is used to identify and count the pixel points in the first adhesive area and the second adhesive area to obtain the actual adhesive area of ​​the insulation board, and calculate the adhesive area ratio of the insulation board in combination with the total area.

9. An intelligent terminal, characterized in that: It comprises a memory and a processor, wherein the memory stores at least one instruction, at least one program, code set or instruction set, and the at least one instruction, at least one program, code set or instruction set is loaded and executed by the processor to implement the insulation board adhesion area ratio detection method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The readable storage medium stores at least one instruction, at least one program, code set or instruction set, and the at least one instruction, at least one program, code set or instruction set is loaded and executed by the processor to implement the insulation board adhesion area ratio detection method as described in any one of claims 1 to 7.

Images