Method and system for segmenting and displaying seam line of stone curtain wall image

Abstract

The invention provides a method and system for segmenting and displaying seam lines of a stone curtain wall image, and the method comprises the steps: scanning a stone curtain wall through millimeter wave equipment, and obtaining a millimeter wave image of the whole wall; identifying a seam line in the millimeter wave image, and recording the coordinate position of the seam line; according to the coordinates of the seam line, segmenting the millimeter wave image of the whole wall into single curtain wall images, and recording coordinate data of each curtain wall; using the coordinate data of the single curtain wall to regenerate an image of the single curtain wall; naming the segmented single curtain wall image by coordinates, and storing the single curtain wall image in a specified folder; and displaying the segmented single curtain wall image in the form of a square overview on a software interface, and supporting a user to check the detailed image of the single curtain wall. According to the invention, through automatic identification and image segmentation of the butt joint seam, a user can directly check a detailed image of a single curtain wall, the trouble of searching a specific area in the whole wall image is avoided, and the detection efficiency is greatly improved.

Description

Technical Field

[0001] The invention relates to the technical field of building curtain wall detection, and in particular to a method and a system for segmenting and displaying seam lines of a stone curtain wall image. Background Art

[0002] Stone curtain walls are widely used in modern buildings due to their beauty and durability. In order to ensure the safety and quality of stone curtain walls, they usually need to be inspected regularly. Millimeter wave imaging is one of the effective methods for perspective inspection of stone curtain walls. Through millimeter wave imaging technology, it is possible to penetrate the stone curtain wall and obtain images of its internal structure, thereby detecting the status of components such as pendants and keels behind the curtain wall.

[0003] At present, the millimeter wave image data display of stone curtain walls mostly adopts the image display method of the whole wall; users can view the detailed image of a certain part through the image of the whole wall, or view the data around a certain coordinate through the coordinate method; although this method can provide an overall view, it has some shortcomings in practical applications.

[0004] Deficiencies of existing technology: 1. Data redundancy: The current millimeter wave image viewing method easily leads to redundant data; the image of the entire wall contains a large amount of information, and users need to spend more time and energy to find the area of ​​concern when viewing, which affects the detection efficiency.

[0005] 2. Lack of precision: Users' viewing needs for curtain wall images are usually based on a piece of curtain wall rather than the entire wall; existing technologies cannot easily and accurately divide the image of the entire wall into a single piece of curtain wall image, making it difficult for users to accurately locate and view the specific situation of a piece of curtain wall.

[0006] 3. Difficulty in follow-up: Existing technology is not convenient for subsequent tracking and searching; once a problem is found in a curtain wall, it is difficult to quickly locate the specific location, which increases the difficulty of repair and maintenance.

[0007] 4. Image quality issues: If a single curtain wall fixture is located at the edge of a single short-distance image during long-distance scanning, the fixture may not be clearly displayed due to image blur or being covered by artifacts; this will affect the accuracy and reliability of the detection results.

[0008] 5. Poor User Experience: The current image display method is not intuitive enough. Users need to manually search and compare images of different areas, which increases the complexity of operations and user burden. The lack of a friendly user interface and interactive design results in a poor user experience during use.

[0009] Therefore, the prior art has deficiencies and needs further improvement. Summary of the invention

[0010] In view of the problems existing in the prior art, the present invention provides a method and system for segmenting and displaying seam lines of a stone curtain wall image.

[0011] To achieve the above object, the specific scheme of the present invention is as follows: The present invention provides a system for segmenting seam lines of stone curtain wall images, the system comprising: An image acquisition module is used to scan the stone curtain wall using a millimeter wave device to obtain a millimeter wave image of the entire wall; A seam line recognition module is used to identify straight lines with higher grayscale values ​​in the millimeter wave image as seam lines, and record the coordinate positions of these seam lines; An image segmentation module is used to segment the millimeter wave image of the entire wall into individual curtain wall images according to the identified seam line coordinates, and record the coordinate data of each curtain wall; A re-imaging module, used to regenerate an image of a single curtain wall from original millimeter wave data using coordinate data of the image of the single curtain wall; The image naming and storage module is used to name the segmented single curtain wall image with coordinates and store it in a specified folder; A display module is used to display the segmented single-block curtain wall image in the form of a block overview on the software interface, and supports users to view the detailed image of the single-block curtain wall image; The user interaction module is used to support users to view the detailed image of a single curtain wall by clicking on the overview image, and supports operations such as zooming in and out.

[0012] Furthermore, the seam line recognition module adopts an artificial intelligence recognition algorithm, which can accurately identify the curtain wall seam line and record the coordinates of the lower left corner and the upper right corner of the seam line.

[0013] Furthermore, when the image segmentation module generates an image of a single curtain wall, it will expand the image outward by 5-10 cm to ensure the clarity of the hanging parts or joints.

[0014] Furthermore, when generating a single curtain wall image, the image segmentation module automatically detects and corrects image distortion caused by scanning angles or environmental factors.

[0015] Furthermore, the display module allows the user to view a detailed image of a single curtain wall by clicking on the block overview image.

[0016] The present invention also provides a method for segmenting seam lines of a stone curtain wall image. Based on the above system, the method comprises the following steps: S1, use millimeter wave equipment to scan the stone curtain wall and obtain the millimeter wave image of the entire wall; S2, identifying straight lines with higher grayscale values ​​in the millimeter wave image as seam lines, and recording the coordinate positions of these seam lines; S3, according to the identified seam line coordinates, the millimeter wave image of the entire wall is divided into single curtain wall images, and the coordinate data of each curtain wall is recorded; S4, using the coordinate data of the single curtain wall, regenerates the image of the single curtain wall from the original millimeter wave data; S5, naming the segmented single curtain wall image according to the coordinates and storing it in a designated folder; S6, displays the segmented single curtain wall image in the form of a block overview on the software interface, and supports users to view the detailed image of a single curtain wall.

[0017] Furthermore, in step S2, in the step of identifying the seam line, the artificial intelligence recognition algorithm used can accurately confirm the position of each curtain wall relative to the wall surface.

[0018] Furthermore, in step S4, in the step of regenerating the image of a single curtain wall, imaging is performed with an expansion of 5-10 cm on both sides of the seam line to ensure the integrity and clarity of the image.

[0019] Furthermore, in step S6, in the step of displaying the single curtain wall image, the single curtain wall image is arranged and displayed according to its actual position on the wall surface, so as to facilitate the user to track and find the problematic stone block.

[0020] Furthermore, in step S6, the user can quickly locate and view a specific single curtain wall image by clicking on the overview image, and zoom-in and zoom-out operations are supported, thereby improving the efficiency and accuracy of curtain wall inspection.

[0021] The technical solution of the present invention has the following beneficial effects: 1. Improve detection efficiency: Through automatic recognition and image segmentation of seams, users can directly view detailed images of a single curtain wall, avoiding the trouble of searching for specific areas in the entire wall image, greatly improving detection efficiency.

[0022] 2. Enhanced accuracy: The generation of single curtain wall images ensures the independence and integrity of each curtain wall unit, especially in the processing of hanging parts or joints. The external expansion imaging ensures the clarity of these key parts and improves the accuracy of detection.

[0023] 3. Facilitate follow-up tracking: Each curtain wall image has clear coordinate information. Once a problem is found, the specific curtain wall position can be quickly located, which simplifies the repair and maintenance workflow and reduces costs.

[0024] 4. Improve image quality: The regenerated single-block curtain wall image not only maintains high resolution, but also reduces the impact of edge effects through external expansion imaging technology, making the image clearer, reducing the interference of artifacts, and improving image quality.

[0025] 5. Optimize user experience: It provides an intuitive block overview and detailed single-block curtain wall image display, and users can view the required information with simple clicks. It also supports advanced functions such as zooming in and out, greatly improving the user's operating experience.

[0026] 6. Reduce data redundancy: The segmented single-block curtain wall image only contains necessary information, avoiding the large amount of redundant data brought by the entire wall image, allowing users to focus more on inspecting key areas and improve work concentration.

[0027] 7. Improve detection reliability: Through precise image processing and display, misjudgments caused by human factors are reduced and the reliability and accuracy of detection results are improved.

[0028] 8. Strong adaptability: This method is suitable for stone curtain walls of different sizes and types, has strong versatility and adaptability, and can be widely used in the inspection of various construction projects. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 is the curtain wall image collected by the present invention; Figure 2 It is a coordinate schematic diagram of a single curtain wall of the present invention; Figure 3 It is a schematic diagram of the present invention showing a single curtain wall image in the form of a block overview diagram; Figure 4 It is the millimeter wave image of the whole wall of the present invention; Figure 5 It is a millimeter wave image of a single curtain wall after seam identification and segmentation according to the present invention; Figure 6 It is a system principle block diagram of the present invention; Figure 7 It is the overall flow chart of the present invention. DETAILED DESCRIPTION

[0030] The present invention is further described in detail below in conjunction with the accompanying drawings and embodiments; it can be understood that the specific embodiments described herein are only used to explain the present invention rather than to limit the present invention; it should also be noted that, for ease of description, only the parts related to the present invention rather than all are shown in the accompanying drawings.

[0031] Combination Figure 1-Figure 7 As shown, the present invention provides a method and system for segmenting and displaying seam lines of a stone curtain wall image, the system comprising: An image acquisition module is used to scan the stone curtain wall using a millimeter wave device to obtain a millimeter wave image of the entire wall; A seam line recognition module is used to identify straight lines with higher grayscale values ​​in the millimeter wave image as seam lines, and record the coordinate positions of these seam lines; An image segmentation module is used to segment the millimeter wave image of the entire wall into individual curtain wall images according to the identified seam line coordinates, and record the coordinate data of each curtain wall; A re-imaging module, used to regenerate an image of a single curtain wall from original millimeter wave data using coordinate data of the image of the single curtain wall; The image naming and storage module is used to name the segmented single curtain wall image with coordinates and store it in a specified folder; A display module is used to display the segmented single-block curtain wall image in the form of a block overview on the software interface, and supports users to view the detailed image of the single-block curtain wall image; The user interaction module is used to support users to view the detailed image of a single curtain wall by clicking on the overview image, and supports operations such as zooming in and out.

[0032] The seam line recognition module adopts an artificial intelligence recognition algorithm, which can accurately identify the curtain wall seam line and record the coordinates of the lower left corner and the upper right corner of the seam line.

[0033] When the image segmentation module generates an image of a single curtain wall, it will expand 5-10 cm outward for imaging to ensure the clarity of the hanging parts or joints.

[0034] When generating a single curtain wall image, the image segmentation module automatically detects and corrects image distortion caused by scanning angles or environmental factors.

[0035] The display module allows the user to view a detailed image of a single curtain wall by clicking on the block overview image.

[0036] The present invention also provides a method for segmenting seam lines of a stone curtain wall image. Based on the above system, the method comprises the following steps: S1, use millimeter wave equipment to scan the stone curtain wall and obtain the millimeter wave image of the entire wall; S2, identifying straight lines with higher grayscale values ​​in the millimeter wave image as seam lines, and recording the coordinate positions of these seam lines; S3, according to the identified seam line coordinates, the millimeter wave image of the entire wall is divided into single curtain wall images, and the coordinate data of each curtain wall is recorded; S4, using the coordinate data of the single curtain wall, regenerates the image of the single curtain wall from the original millimeter wave data; S5, naming the segmented single curtain wall image according to the coordinates and storing it in a designated folder; S6, displays the segmented single curtain wall image in the form of a block overview on the software interface, and supports users to view the detailed image of a single curtain wall.

[0037] In step S2, in the step of identifying the seam line, the artificial intelligence recognition algorithm used can accurately confirm the position of each curtain wall relative to the wall surface.

[0038] In step S4, in the step of regenerating the image of a single curtain wall, imaging is performed by expanding 5-10 cm on both sides of the seam line to ensure the integrity and clarity of the image.

[0039] In step S6, in the step of displaying the single curtain wall image, the single curtain wall image is arranged and displayed according to its actual position on the wall surface, so as to facilitate the user to track and find the problematic stone block.

[0040] In step S6, the user quickly locates and views a specific single curtain wall image by clicking on the overview image, and supports zooming in and out operations, thereby improving the efficiency and accuracy of curtain wall inspection.

[0041] Working principle of the present invention The present invention proposes a method and system for segmenting and displaying the seam lines of a stone curtain wall image. The working principle mainly includes the following steps: Millimeter wave image acquisition: Use millimeter wave equipment to scan the stone curtain wall and obtain the millimeter wave image of the entire wall.

[0042] Millimeter wave equipment can penetrate the stone curtain wall and obtain images of its internal structure, including the status of components such as pendants and keels. These image data are the basis for subsequent processing.

[0043] Seam line identification: Using artificial intelligence recognition algorithms, straight lines with higher grayscale values ​​in millimeter wave images are identified as seam lines, and the coordinate positions of these seam lines are recorded.

[0044] Through deep learning or other artificial intelligence algorithms, the straight line features in the image are identified, especially the parts with higher grayscale values. These parts are highly likely to be curtain wall seams. The coordinate positions of these seams are recorded, including the coordinates of the lower left corner and the upper right corner, so as to determine the boundaries of each curtain wall.

[0045] Image Segmentation: According to the identified seam line coordinates, the millimeter wave image of the entire wall is segmented into single curtain wall images, and the coordinate data of each curtain wall is recorded.

[0046] The coordinate position of the seam line is used to divide the image of the entire wall into multiple independent single-block curtain wall images. When generating a single-block curtain wall image, the image will be expanded 5-10cm to ensure the clarity of the hanging parts or seams and avoid image blur.

[0047] Re-imaging: The coordinate data of the single curtain wall is used to regenerate the image of the single curtain wall from the original millimeter wave data.

[0048] According to the coordinate data of each curtain wall, the data of the corresponding area is extracted from the original millimeter wave data for re-imaging. This step ensures the integrity and clarity of the image of a single curtain wall and avoids image distortion caused by scanning angles or environmental factors.

[0049] Image naming and storage: The segmented single curtain wall images are named according to coordinates and stored in the specified folder.

[0050] Each curtain wall image is named according to its coordinate position on the wall, which is convenient for subsequent position management and image data tracking. It is stored in a designated folder for easy management and call.

[0051] Image display: The software interface displays the segmented single curtain wall image in the form of a block overview diagram, and supports users to view the detailed image of a single curtain wall.

[0052] According to the size and coordinate position of each curtain wall, the segmented single curtain wall image is displayed in the form of a square overview on the software interface. Users can click on the overview to view the detailed image of a single curtain wall, which supports zooming in and out, providing intuitive visual effects.

[0053] User Interaction: Users can click on the overview image to view a detailed image of a single curtain wall, which supports operations such as zooming in and out.

[0054] It provides a friendly user interface and interactive design, allowing users to view the required detailed images through simple clicks. It supports multiple operation modes to improve the user's operating experience.

[0055] Workflow Summary Millimeter wave image acquisition: Get millimeter wave images of the entire wall.

[0056] Seamline Recognition: Identify seamlines in an image and record their coordinates.

[0057] Image segmentation: segment the curtain wall image into individual pieces according to the seam line coordinates.

[0058] Re-imaging: Generate a high-resolution image of a single curtain wall from the raw data.

[0059] Image naming and storage: Name and store single curtain wall images with coordinates.

[0060] Image display: Display block overview and detailed images on the software interface.

[0061] User interaction: Supports users to view detailed images by clicking on the overview image, and supports multiple operation modes.

[0062] Through the above steps, the present invention effectively solves the problems of redundant data, insufficient accuracy, difficulty in follow-up tracking, poor image quality and user experience in the prior art, and provides an efficient, accurate and easy-to-use stone curtain wall detection method.

[0063] Embodiment 1: Millimeter wave imaging of stone curtain walls is an effective method for perspective detection of curtain walls. Currently, data is mostly displayed using millimeter wave images of the entire wall. You can view detailed images of a certain part separately, check the status of the hangers and keels behind the curtain wall, or view data around a certain coordinate in coordinate mode.

[0064] The current millimeter wave image viewing method easily leads to redundant data. Users' needs for viewing curtain wall images should be based on a piece of curtain wall. There is no need to view the image data of the entire wall. In addition, the current technology is not convenient for subsequent tracking and searching, and it is impossible to accurately locate which piece of curtain wall has a problem.

[0065] Based on this, in order to solve the above problem that it is impossible to view the accurate curtain wall image, a method for segmenting and displaying the seam lines of the millimeter wave stone curtain wall image is provided.

[0066] This system provides the following technical solutions: the system includes millimeter wave image seam line recognition, millimeter wave image segmentation, and curtain wall image display after segmentation.

[0067] The newly developed technology is a supporting software that will use recognition algorithms, imaging algorithms and software UI interfaces to work together to segment and display the millimeter wave stone curtain wall image seams.

[0068] The first step is to use artificial intelligence recognition algorithms to identify the straight lines and high grayscale values ​​in the curtain wall image. This part is very likely to be the curtain wall seam line. Record the coordinates of these seams. The enclosed part is the entire curtain wall. Record the coordinates of the lower left corner and upper right corner of the curtain wall.

[0069] The second step is to stitch all the currently acquired millimeter wave images together, and re-image the millimeter wave raw data through the acquired curtain wall coordinates to generate the image data of a single curtain wall. The coordinate data M_X1, M_X2, M_Y1, and M_Y2 of this curtain wall are recorded, and the curtain wall image is named with the coordinates to facilitate subsequent position management.

[0070] like Figure 2 As shown in the figure, the meanings of M, N, X, and Y are as follows: M: represents the number of rows of the millimeter wave image (vertical direction); N: represents the number of columns of the millimeter wave image (horizontally); X: represents the horizontal coordinate of the millimeter wave image; Y: represents the vertical coordinate of the millimeter wave image; Specifically, in the picture: M_0 and M_(N-1) are the coordinates representing the top and bottom boundaries of the millimeter wave image; M_XL and M_YL are the coordinates representing the lower left corner of a single curtain wall image; M_XR and M_YR are the coordinates representing the upper right corner of a single curtain wall image; M_X1 and M_Y1 are the coordinates representing the upper left corner of a single curtain wall image; M_X2 and M_Y2 are the coordinates representing the lower right corner of a single curtain wall image; These coordinates are used to define the position of a single curtain wall image in the entire millimeter wave image.

[0071] The third step is to display the single-block millimeter-wave curtain wall image after segmentation in the previous step in the software. If the range of a task is: the starting point is (0,0) and the end point is (200,300), the display area ratio of the software needs to be adjusted to 200:300, which is consistent with the set scanning range. The lower left corner is the origin, and the curtain wall image is displayed in the form of a square overview diagram on the completion page. The position of each curtain wall needs to be accurately displayed according to the size and coordinates of the curtain wall.

[0072] Click on an overview image to view a detailed image of this curtain wall.

[0073] It has the following advantages: (1) Artificial intelligence methods are used to identify curtain wall joints, which can accurately determine the position of each curtain wall relative to the wall surface.

[0074] (2) If the hanging parts of a single curtain wall are located at the edge of a single short-distance image during long-distance scanning, the hanging parts may be blurred or covered by artifacts. Therefore, it is necessary to expand the image 5-10 cm outward. After obtaining the edge seam position of a single curtain wall, re-image using the original data at the edge seam position can avoid this problem and improve data integrity.

[0075] (3) The millimeter wave image of the entire curtain wall is displayed in the form of an overview diagram, which very intuitively shows the arrangement of the stones on the wall. If problems are found later, the problematic stone blocks can be quickly traced.

[0076] The actual process of seam identification: The diagonal coordinates of the first tile are (25, 136), (80, 356)! The diagonal coordinates of the second tile are (60, 136), (121, 356)! The diagonal coordinates of the 3rd tile are (101, 136), (193, 356)! The diagonal coordinates of the 4th tile are (173, 136), (215, 356)! The diagonal coordinates of the 5th tile are (195, 136), (254, 356)! The diagonal coordinates of the second tile are (25, 336), (80, 555)! The diagonal coordinates of the 4th tile are (60, 336), (121, 555)! The diagonal coordinates of the 6th tile are (101, 336), (193, 555)! The diagonal coordinates of the 8th tile are (173, 336), (215, 555)! The diagonal coordinates of the 10th tile are (195, 336), (254, 555)! Found 4 split image files! d:\aidata\Task_20240912172623\5_Edge\RAWDATA12282890_2_800_200_4_0_0_BiNpath_imgxy1.crd file size is 140 bytes The number of seam coordinates is 5; The center coordinates of the gap rectangle are (123, 152), the rectangle is 245 wide and 15 high. It is in the first segmentation image. The number of rows in the segmented image is 200, and the correct probability of the gap is 87.371185! The center coordinates of the gap rectangle are (35, 102), the rectangle is 16 wide and 203 high. It is in the first segmentation image. The number of rows in the segmented image is 200, and the correct probability of the gap is 74.667274! The center coordinates of the gap rectangle are (183, 101), the rectangle is 15 wide and 193 high. It is in the first segmentation image. The number of rows in the segmented image is 200, and the correct probability of the gap is 69.502693! The center coordinates of the gap rectangle are (111, 98), the rectangle is 16 wide and 187 high. It is in the first segmentation image. The number of rows in the segmented image is 200, and the correct probability of the gap is 66.615440! The center coordinates of the gap rectangle are (246, 97), the rectangle is 15 wide and 193 high. It is in the first segmentation image. The number of rows in the segmented image is 200, and the correct probability of the gap is 58.968204! The 0th gap is a horizontal gap, and its y coordinate in the whole picture is 152! The first gap is a vertical gap, and its x-coordinate in the whole picture is 35! The second gap is a vertical gap, and its x-coordinate in the whole picture is 183! The third gap is a vertical gap, and its x-coordinate in the whole picture is 111! The fourth gap is a vertical gap, and its x-coordinate in the whole picture is 246! d:\aidata\Task_20240912172623\5_Edge\RAWDATA12282890_2_800_200_4_1_0_0_BiNpath_imgxy1.crd file size is 140 bytes The number of seam coordinates is 5 The center coordinates of the gap rectangle are (126, 152), the rectangle width is 251, the height is 16, it is in the second segmentation image, the number of rows of the segmented image is 200, and the correct probability of the gap is 87.042595! The center coordinates of the gap rectangle are (34, 99), the rectangle is 14 wide and 197 high. It is in the second segmentation image. The number of rows in the segmented image is 200, and the correct probability of the gap is 74.673424! The center coordinates of the gap rectangle are (184, 101), the rectangle is 16 wide and 201 high. It is in the second segmentation image. The number of rows in the segmented image is 200, and the correct probability of the gap is 61.719334! The center coordinates of the gap rectangle are (205, 102), the rectangle is 15 wide and 200 high. It is in the second segmentation image. The number of rows in the segmented image is 200, and the correct probability of the gap is 52.664364! The center coordinates of the gap rectangle are (246, 97), the rectangle is 15 wide and 193 high. It is in the second segmentation image. The number of rows in the segmented image is 200, and the correct probability of the gap is 58.968204! The 0th gap is a horizontal gap, and its y coordinate in the whole picture is 352! The first gap is a vertical gap, and its x-coordinate in the whole picture is 34! The second gap is a vertical gap, and its x-coordinate in the whole picture is 184! The third gap is a vertical gap, and its x-coordinate in the whole picture is 205! The fourth gap is a vertical gap, and its x-coordinate in the whole picture is 246! d:\aidata\Task_20240912172623\5_Edge\RAWDATA12282890_2_800_200_4_2_0_0_BiNpath_imgxy1.crd file size is 168 bytes The number of seam coordinates is 6; The center coordinates of the gap rectangle are (118, 152), the rectangle width is 237, the height is 16, it is in the third segmentation image, the number of rows of the segmented image is 200, and the correct probability of the gap is 83.141785! The center coordinates of the gap rectangle are (34, 99), the rectangle is 14 wide and 197 high. It is in the third segmentation image. The number of rows in the segmented image is 200, and the correct probability of the gap is 74.673424! The center coordinates of the gap rectangle are (184, 101), the rectangle is 16 wide and 201 high. It is in the third segmentation image. The number of rows in the segmented image is 200, and the correct probability of the gap is 61.719334! The center coordinates of the gap rectangle are (205, 102), the rectangle is 15 wide and 200 high. It is in the third segmentation image. The number of rows in the segmented image is 200, and the correct probability of the gap is 52.664364! The center coordinates of the gap rectangle are (246, 97), the rectangle is 15 wide and 193 high. It is in the third segmentation image. The number of rows in the segmented image is 200, and the correct probability of the gap is 58.968204! The center coordinates of the gap rectangle are (276, 97), the rectangle is 15 wide and 193 high. It is in the third segmentation image. The number of rows in the segmented image is 200, and the correct probability of the gap is 58.968204! The 0th gap is a horizontal gap, and its y coordinate in the whole picture is 352! The first gap is a vertical gap, and its x-coordinate in the whole picture is 34! The second gap is a vertical gap, and its x-coordinate in the whole picture is 184! The third gap is a vertical gap, and its x-coordinate in the whole picture is 205! The fourth gap is a vertical gap, and its x-coordinate in the whole picture is 246! The fifth gap is a vertical gap, and its x-coordinate in the whole picture is 276! This method is implemented in a newly developed curtain wall detection system, and the curtain wall is scanned by a millimeter wave device to obtain a millimeter wave image of the entire surface.

[0077] The front millimeter wave image is stored, and this method is used to identify the joints of the millimeter wave image of the entire wall to obtain the coordinate positions of all the stones on this wall. The specific method is as follows: Wall_edge_detect Identify the seams of multiple image files in a folder and output a file containing the coordinate information of each curtain wall. Multiple image files should be the same size. Wall_edge_detect(int single_row,float connf,string input_image_path, string outpu_edge_path,string out_all_coor_path) single_row single file row number connf identification threshold input_image_path Enter the path to the image folder of the mmWave building curtain wall outpu_edge_path Output the folder path for labeling the position information of each single curtain wall out_all_coor_path Outputs the coordinate information of the individual bricks that make up the entire image. The input path is the full path For example, when performing a long scan, real-time imaging is performed every 1m, and 8 files are output each time. Suppose the file is named The images output by the first imaging are: 1_Imgxy1, 1_Imgxy2, ..., 1_Imgxy6, 1_Imgxz, 1_Imgyz The images output by the second imaging are: 2_Imgxy1, 2_Imgxy2, ..., 2_Imgxy6, 2_Imgxz, 2_Imgyz ………… The images output by the Nth imaging are: N_Imgxy1, N_Imgxy2, ..., N_Imgxy6, N_Imgxz, N_Imgyz Then 1_Imgxy1, 2_Imgxy1, ..., N_Imgxy1 (all of which are front views of the curtain wall surface layer) should be placed in the input image folder. Wall_edge_detect will read these front view files in the order of the file names, identify the seams and segment the position information of each single curtain wall in the long image composed of these multiple images. Output the file containing the position information of a single curtain wall in the following format: path. The coordinate format of the output file is: Coordinate format: x1, y1, x2, y2 File format: hexadecimal .txt file; After obtaining the coordinates of a single curtain wall, use the coordinates of the single curtain wall for re-imaging to avoid image blur at the joints. Obtain a single curtain wall image, record the obtained x1, y1, x2, y2 coordinates, and bind the coordinate data of the curtain wall.

[0078] After the scanning is completed, the segmented single curtain wall images are arranged and displayed.

[0079] Example 2: Stone curtain wall maintenance inspection background A large commercial complex needs to perform regular maintenance inspections on its exterior stone curtain wall to ensure the safety and aesthetics of the curtain wall. Traditional inspection methods are time-consuming and labor-intensive, and it is difficult to find hidden problems. In order to improve efficiency and accuracy, the complex adopts an inspection system based on the technology of the present invention.

[0080] Implementation process Millimeter-wave image acquisition: Technicians use drones equipped with millimeter-wave equipment to fly along the building's exterior wall, scanning the entire stone curtain wall and collecting high-quality millimeter-wave image data.

[0081] Seam line identification: The collected millimeter wave images are uploaded to the cloud server, which runs a pre-trained artificial intelligence model to automatically identify all seams and mark their coordinates.

[0082] Image segmentation and re-imaging: Based on the coordinates of the seam lines, the system automatically segments the image of the entire wall into individual curtain wall images and regenerates these individual images from the original millimeter wave data, ensuring image clarity.

[0083] Image display and analysis: On a dedicated software interface, technicians can view overviews and detailed images of all individual curtain walls. By comparing images taken at different time points, technicians can quickly locate areas where potential hazards may exist, such as cracks, hollows, etc.

[0084] Report generation and maintenance recommendations: The system automatically generates a detailed inspection report, including the location, severity, and recommended maintenance measures for each suspected problem. The technicians develop a specific maintenance plan based on the report.

[0085] Example 3: Stone curtain wall installation quality control background After a construction company completed the installation of a stone curtain wall in a high-end residential area, it needed to strictly control the installation quality to ensure that no potential quality problems were missed. To this end, the company decided to use the technical means provided by the present invention to perform quality control.

[0086] Implementation process Millimeter wave image acquisition: After the curtain wall was installed, the project team used a portable millimeter wave scanner to conduct a comprehensive scan of the entire stone curtain wall to ensure that every corner was covered.

[0087] Seam line identification and verification: By uploading the scanned millimeter wave image to the company's data center, the AI ​​algorithm is used to automatically identify and verify whether the position of the seam line meets the requirements of the design drawings, and check whether there are any abnormalities.

[0088] Image segmentation and detail inspection: The system automatically completes the image segmentation and re-imaging, and project team members can check the installation quality of each stone curtain wall piece by piece, paying special attention to the tightness and flatness of the joints.

[0089] Problem marking and rectification: For problems found during the inspection, such as uneven joints and scratches on the stone surface, the project team will mark them in the system and arrange for professionals to make timely rectifications.

[0090] Final acceptance and delivery: After all rectification work is completed, comprehensive millimeter wave image acquisition and inspection are carried out again to ensure that all problems have been properly resolved. Subsequently, the final acceptance report is prepared and the high-quality stone curtain wall project is delivered to the owner.

[0091] The above two embodiments demonstrate the flexibility and efficiency of the technology of the present invention in practical applications. It is not only suitable for the maintenance of existing stone curtain walls, but can also play an important role in the construction phase of new projects.

[0092] The above description is only a preferred embodiment of the present invention, and does not limit the scope of the present invention. All equivalent structural changes made by using the contents of the present invention specification and drawings under the inventive concept of the present invention, or directly / indirectly applied in other related technical fields are included in the protection scope of the present invention.

Claims

1. A system for segmenting seam lines of stone curtain wall images, characterized in that: The system comprises: An image acquisition module is used to scan the stone curtain wall using a millimeter wave device to obtain a millimeter wave image of the entire wall; A seam line recognition module is used to identify straight lines with higher grayscale values ​​in the millimeter wave image as seam lines, and record the coordinate positions of these seam lines; An image segmentation module is used to segment the millimeter wave image of the entire wall into individual curtain wall images according to the identified seam line coordinates, and record the coordinate data of each curtain wall; A re-imaging module, used to regenerate an image of a single curtain wall from original millimeter wave data using coordinate data of the image of the single curtain wall; The image naming and storage module is used to name the segmented single curtain wall image with coordinates and store it in a specified folder; A display module is used to display the segmented single-block curtain wall image in the form of a block overview on the software interface, and supports users to view the detailed image of the single-block curtain wall image; The user interaction module is used to support users to view the detailed image of a single curtain wall by clicking on the overview image, and supports operations such as zooming in and out.

2. The system according to claim 1, characterized in that The seam line recognition module adopts an artificial intelligence recognition algorithm, which can accurately identify the curtain wall seam line and record the coordinates of the lower left corner and the upper right corner of the seam line.

3. The system according to claim 1, characterized in that When the image segmentation module generates an image of a single curtain wall, it will expand 5-10 cm outward for imaging to ensure the clarity of the hanging parts or joints.

4. The system according to claim 1, characterized in that When generating a single curtain wall image, the image segmentation module automatically detects and corrects image distortion caused by scanning angles or environmental factors.

5. The system according to claim 1, characterized in that The display module allows the user to view a detailed image of a single curtain wall by clicking on the block overview image.

6. A method for segmenting seam lines of a stone curtain wall image, based on the system according to any one of claims 1 to 5, characterized in that: The method comprises the following steps: S1, use millimeter wave equipment to scan the stone curtain wall and obtain the millimeter wave image of the entire wall; S2, identifying straight lines with higher grayscale values ​​in the millimeter wave image as seam lines, and recording the coordinate positions of these seam lines; S3, according to the identified seam line coordinates, the millimeter wave image of the entire wall is divided into single curtain wall images, and the coordinate data of each curtain wall is recorded; S4, using the coordinate data of the single curtain wall, regenerates the image of the single curtain wall from the original millimeter wave data; S5, naming the segmented single curtain wall image according to the coordinates and storing it in a designated folder; S6, displays the segmented single curtain wall image in the form of a block overview on the software interface, and supports users to view the detailed image of a single curtain wall.

7. The method according to claim 6, characterized in that In step S2, in the step of identifying the seam line, the artificial intelligence recognition algorithm used can accurately confirm the position of each curtain wall relative to the wall surface.

8. The method according to claim 6, characterized in that In step S4, in the step of regenerating the image of a single curtain wall, imaging is performed by expanding 5-10 cm on both sides of the seam line to ensure the integrity and clarity of the image.

9. The method according to claim 6, characterized in that In step S6, in the step of displaying the single curtain wall image, the single curtain wall image is arranged and displayed according to its actual position on the wall surface, so as to facilitate the user to track and find the problematic stone block.

10. The method according to claim 6, characterized in that In step S6, the user quickly locates and views a specific single curtain wall image by clicking on the overview image, and supports zooming in and out operations, thereby improving the efficiency and accuracy of curtain wall inspection.

Images