Large-area continuous engineering facade intelligent inspection grid positioning and data management method

Abstract

This invention discloses an intelligent inspection grid positioning and data management method for large-area continuous engineering facades, relating to the field of engineering safety monitoring and inspection technology. It is particularly applicable to engineering facades with large areas, continuous surfaces, and lacking significant feature points, such as dam faces and engineering slope surfaces. The method includes: firstly, unified management of positioning and image data collected by intelligent inspection equipment, mapping the collected data onto an orthophoto map of the large-area continuous engineering facade to facilitate accurate location and tracking of defects during inspection; secondly, a grid-based representation method for surface defect data of large-area continuous engineering facades is proposed, establishing a grid-based defect data representation model for large-area continuous engineering facades to achieve standardized recording and efficient management of defect information.

Description

Technical Field

[0001] This invention relates to the field of safety monitoring and maintenance technology for large-area continuous engineering facades, and particularly to a method for intelligent inspection grid positioning and data management of large-area continuous engineering facades using intelligent inspection equipment. Background Technology

[0002] In the safety monitoring and operation and maintenance of tall engineering buildings, the detection and tracing of surface defects such as cracks has always been a key issue. Intelligent inspection equipment (such as wall-climbing robots), as an emerging close-range inspection tool, can directly attach to the facades of large, continuous engineering projects to conduct detailed inspections. Compared with drones or manual inspections, it has many advantages, including high detection accuracy, strong close-range perception capabilities, low operational safety risks, and high inspection efficiency. However, for large-scale tall structures, such as dams, which can reach widths of 800 meters and heights of 300 meters, with vast facade areas, the positioning accuracy of intelligent inspection equipment on the continuous facades of large, tall buildings is greatly affected by signal obstruction from the tall structures, errors caused by multiple reflections in narrow valleys, and accumulated errors in inertial navigation devices. How to spatially locate defects during the inspection of large and continuous engineering facades and simultaneously match positioning data with image data is a pressing technical challenge.

[0003] In existing large-area continuous engineering facade inspections, the local positioning coordinates of intelligent inspection equipment are typically used to record defect locations. However, this coordinate system cannot directly correspond to the overall coordinate system of the large-area continuous engineering facade, making it difficult to trace defect data across the entire facade. This is especially true for dams, which are divided vertically into multiple elevations and horizontally into different inspection unit areas. (For example, a common division method for dams is to treat a dam section as a single inspection unit, with each section approximately 18–20 meters wide and 30–50 meters high.) Existing methods based on the local positioning coordinates of intelligent inspection equipment have limited accuracy and cannot meet the requirements for meter-level location and tracing of defects such as cracks and spalling.

[0004] On the other hand, large-scale continuous facade inspections not only require the collection and recording of defects, but also the accurate marking of "defective" and "defect-free" locations so that maintenance personnel can quickly identify safety risk areas and formulate maintenance plans. However, the existing methods of expressing inspection data are fragmented and lack unified grid-based planning and management methods, which often brings difficulties to subsequent data queries, defect tracking and investigation, and defect elimination operations. Summary of the Invention

[0005] The main objective of this invention is to propose a method for intelligent inspection grid positioning and data management of large-area continuous engineering facades. This method addresses the problem in existing technologies where intelligent inspection equipment struggles to uniformly collect and effectively trace defect data during the inspection of large-area continuous engineering facades. It can match inspection data with the overall coordinate system of large-area continuous engineering facades, achieving standardized labeling and unified management of defect locations, thus facilitating efficient troubleshooting and safety diagnosis by subsequent maintenance personnel.

[0006] Another objective of this invention is to propose an intelligent inspection grid positioning and data management device for large-area continuous engineering facades.

[0007] To achieve the above objectives, a first aspect of the present invention proposes a method for intelligent inspection grid positioning and data management of large-area continuous engineering facades, comprising:

[0008] Acquire large-area continuous engineering facade images, construct perspective transformation relationships using feature points of the inspection work area shape, perform geometric correction on the original images, and generate orthophotos with true physical size geometric proportions by combining the actual dimensions of the large-area continuous engineering facades. By using the encoder of the winch drive motor installed on the intelligent inspection equipment, combined with the winch diameter, the climbing distance of the intelligent inspection equipment along the suspension rope is calculated, and combined with the start and end points of the large-area continuous engineering facade inspection operation area manually marked, the position of the intelligent inspection equipment is mapped to the pixel coordinate system of the orthophoto. Based on the servo angle of the automatic scanning bracket of the camera mounted on the intelligent inspection equipment, the length of the automatic scanning bracket, and the field of view of the camera, the shooting range of the camera is determined, and the coverage area and orientation of the robot's image are marked in the orthophoto. The orthophotos of large-area continuous engineering facades are divided into horizontal and vertical grids, and a unique naming rule of "elevation-inspection unit-horizontal grid number-vertical grid number" is established to uniformly represent the defect data. When the field of view of the image captured by the intelligent inspection equipment intersects with a certain grid cell, the number of images in that grid is incremented by 1, and the defect is marked in the grid, thereby completing the statistical analysis and visualization of large-area continuous engineering facade defect data based on grid cells.

[0009] To achieve the above objectives, a second aspect of the present invention provides an intelligent inspection grid positioning and data management device for large-area continuous engineering facades, characterized in that it includes: The first module is used to acquire images of large-area continuous engineering facades, construct perspective transformation relationships using feature points of the inspection area of ​​large-area continuous engineering facades, perform geometric correction on the original images, and generate orthophotos with real physical size geometric proportions by combining the actual size of the inspection area of ​​large-area continuous engineering facades. The second module is used to calculate the climbing distance of the smart device along the suspension rope by using an encoder installed on the winch drive motor of the smart device and combining it with the winch diameter, and to combine it with the starting point of the manually calibrated inspection unit area to map the robot position into the pixel coordinate system of the orthophoto. The third module is used to determine the shooting range of the camera based on the rotation angle of the automatic scanning bracket servo of the camera mounted on the intelligent inspection equipment, the length of the automatic scanning bracket, and the field of view parameters of the camera, and to mark the coverage area and orientation of the image captured by the intelligent inspection equipment in the orthophoto. The fourth module is used to divide the orthophotos of the large-area continuous engineering facade inspection area into horizontal and vertical grids, and to establish a unique naming rule of "elevation - inspection unit - horizontal grid number - vertical grid number" to uniformly represent the defect data. The fifth module is used to increment the image count of a grid cell by 1 when the field of view of the image captured by the intelligent inspection equipment intersects with a certain grid cell, and to mark the defect in the grid, thereby completing the statistical analysis and visualization of the appearance defects of a large-area continuous engineering facade based on grid cells.

[0010] The embodiments of the present invention have the following beneficial effects: (1) In the prior art, the inspection data of large-area continuous engineering facades usually rely on manual recording or rough division based on the inspection unit area / elevation, which makes it difficult to achieve fine-grained source tracing of defects. This invention establishes a grid coordinate system based on orthophotos, and maps the positioning data and image data collected by intelligent inspection equipment to the overall coordinate system of large-area continuous engineering facades. It also divides the inspection unit area into meter-level grids in the horizontal and vertical directions, realizing accurate labeling and traceability of defect locations, and significantly improving the efficiency and accuracy of data management.

[0011] (2) Existing technologies mostly rely on a single coordinate system or local reference points, making it difficult to uniformly manage inspection data across different inspection unit areas or different inspection tasks. The grid naming rule of "elevation-inspection unit-horizontal-vertical" proposed in this invention assigns a unique number to each grid unit, ensuring the consistency and comparability of defect data across the entire global area of ​​a large-area continuous engineering facade, thereby facilitating defect comparison analysis and long-term monitoring across time periods and tasks.

[0012] (3) Existing intelligent inspection equipment only records images and lacks unified labeling of images and physical locations. This invention establishes a labeling method for image coverage areas by combining encoder ranging and servo motor angle calculation with camera field of view. By drawing the field of view line and judging its intersection with the grid area, it is possible not only to determine the coverage grid corresponding to the image, but also to count the number of images contained in each grid, realizing the visual management and quantitative analysis of image data.

[0013] (4) The technical solution of this invention can provide maintenance personnel of large-area continuous engineering facades with a clear and intuitive display of inspection results. The defect data is stored and labeled in a grid format, which can quickly locate the damaged area and intuitively reflect the inspection coverage. This avoids the problems of scattered data, difficulty in tracing the source, and low management efficiency in the prior art. Therefore, this invention significantly improves the management efficiency of defect data of large-area continuous engineering facades and the utilization value of inspection results. It has promotion and application value and strong universality. It can be extended to the defect detection and management of other large-area structural surfaces. Attached Figure Description

[0014] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 A flowchart illustrating a method for intelligent inspection grid positioning and data management of large-area continuous engineering facades provided in an embodiment of the present invention; Figure 2 The orthophoto image of the inspection unit area obtained by the transmission transformation of the inspection unit area provided in the embodiment of the present invention is used as the base map for the positioning data annotation. Figure 3 This is a schematic diagram of the positioning structure of the intelligent inspection device provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of installing a multi-turn encoder on a geared motor that drives the winch of a robot, according to an embodiment of the present invention. Figure 5 This is a schematic diagram of the inspection unit area detection grid division provided in an embodiment of the present invention; Figure 6 This is a schematic diagram of the detected damage provided in an embodiment of the present invention; Figure 7 A schematic diagram illustrating the naming rules for the grid coordinate system of large-area continuous engineering facade inspection provided in this embodiment of the invention; Figure 8 Example diagram of image data naming rules provided in embodiments of the present invention; Figure 9 This is a schematic diagram illustrating the practical application of mesh generation and defect characterization provided in an embodiment of the present invention. Detailed Implementation

[0015] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0016] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0017] The following description, with reference to the accompanying drawings, describes a method and apparatus for intelligent inspection grid positioning and data management of large-area continuous engineering facades according to an embodiment of the present invention.

[0018] Example 1 This invention relates to a data grid management method for the inspection of defects on the dam surface using intelligent inspection equipment, such as... Figure 1 As shown, the method includes the following steps: S1. Acquire large-area continuous engineering facade images, construct perspective transformation relationships using feature points of the inspection work area shape, perform geometric correction on the original images, and generate orthophotos with true physical size geometric proportions by combining the actual dimensions of the large-area continuous engineering facade.

[0019] In this embodiment of the invention, firstly, images containing each inspection unit area (taking a wall-climbing robot as an example) are captured using a drone, cropped from an oblique photogrammetry 3D model, or captured by a camera on the ground. The actual height and length of each inspection unit area are known parameters. Then, feature points (such as the coordinates of the four vertices) of the target inspection unit area are selected. Then, the vertices are arranged in order according to the requirements of perspective transformation. Next, the perspective transformation matrix is ​​constructed. Geometric correction is performed on the original image to obtain an orthophoto.

[0020] In one embodiment of the present invention, the perspective transformation relationship is expressed as follows:

[0021] After expansion, we can obtain:

[0022] in, These are the pixel coordinates in the original image. These are the pixel coordinates of the orthophoto after perspective transformation. These are the parameters of the perspective transformation matrix.

[0023] Due to the matrix With 8 degrees of freedom, at least four pairs of points are required to solve the problem. In this invention, the matrix is ​​uniquely determined by establishing a correspondence between the four vertices of the inspection unit region and the four vertices of the target orthorectified rectangle. Through the above transformation process, the image of the inspection unit region can be converted from an oblique viewpoint to an orthorectified viewpoint, thereby obtaining an orthorectified image of the inspection unit region with true geometric proportions.

[0024] In one embodiment of the present invention, after obtaining the orthophoto image, a proportional mapping relationship between pixel coordinates and real physical coordinates is established on the orthophoto image based on the known actual physical size of the inspection unit area, specifically as follows: Let the pixel width of the inspection unit area in the orthophoto be... The pixel height is The actual physical length of the corresponding dam section is The actual physical height is Then the pixel-to-physical-size scaling factors of the orthophoto in the horizontal and vertical directions are respectively:

[0025] in, and These represent the actual physical dimensions of a unit pixel in the horizontal and vertical directions of an orthophoto image, respectively. For any pixel in an orthophoto and Its corresponding position in the actual physical coordinate system It can be represented as

[0026] For the pixel distance between any two pixels in an orthophoto , The corresponding actual physical distance They are respectively

[0027] This enables a quantitative mapping between the pixel scale in the orthophoto and the actual physical scale of the dam surface, allowing the generated orthophoto to simultaneously possess geometric correction characteristics and the actual physical size ratio.

[0028] In one possible embodiment, the orthophoto image of the dam section obtained by the dam section transmission transformation is used as the base map for positioning data annotation, such as... Figure 2 As shown.

[0029] S2 calculates the climbing distance of the intelligent inspection equipment along the suspension rope by using the encoder of the winch drive motor installed on the intelligent inspection equipment and combining it with the winch diameter. It then combines this distance with the start and end points of the large-area continuous engineering facade inspection work area that are manually marked, and maps the position of the intelligent inspection equipment to the pixel coordinate system of the orthophoto.

[0030] In one embodiment of the present invention, the positioning structure of the intelligent inspection device (taking a wall-climbing robot as an example) is as follows: Figure 3 As shown. The intelligent inspection equipment clamps the rope with a winch (9) and pulleys (10) and moves up and down along the rope (8) by relying on the speed reduction motor to drive the winch (9) to rotate. At the same time, the propellers (4) on the left and right sides generate thrust to make the intelligent inspection equipment stably attached to the surface of the large-area continuous engineering facade.

[0031] Due to the obstruction of tall, large-area continuous engineering facades, it is difficult to receive RTK signals, therefore... Figure 4 As shown, in this embodiment of the invention, a multi-turn encoder (6) is installed on the geared motor (7) that drives the winch to rotate. When the geared motor (7) rotates, it drives the multi-turn encoder (6) to rotate via the belt (5), thereby realizing the counting of rotations. Since there is almost no relative slippage between the winch (9) and the rope, the climbing distance of the robot along the rope direction can be calculated by collecting the number of rotations of the motor and combining it with the diameter of the winch.

[0032] The calculation formula is as follows:

[0033] Where d is the climbing distance of the intelligent inspection equipment along the rope, n is the number of motor rotations recorded by the encoder, and D is the diameter of the winch.

[0034] S3 determines the camera's shooting range based on the servo angle of the automatic scanning bracket of the camera mounted on the intelligent inspection equipment, the length of the automatic scanning bracket, and the camera's field of view, and marks the coverage area and orientation of the robot's captured image in the orthophoto.

[0035] In this embodiment of the invention, the intelligent inspection device is equipped with an automatic scanning bracket on top. The automatic scanning bracket is driven by a servo motor, and the servo motor controller can output a target rotation angle command to determine the angle of the automatic scanning bracket. A camera with a fixed field of view is installed at the end of the automatic scanning bracket.

[0036] Given the servo motor angle θ, the length L of the automatic scanning bracket, and the camera's field of view φ, the relative position of the area covered by the camera's image to the robot body can be determined. Its coverage width range can be expressed as:

[0037] Where W represents the coverage width of the camera on the dam surface.

[0038] When combining the physical scale of orthophotos, this invention utilizes a starting point calibration method, taking the initial position of the intelligent inspection device in the orthophoto of the inspection unit area as a reference point. Subsequently, based on the vertical climbing distance *d* calculated by the winch encoder, the position of the intelligent inspection device is mapped to the corresponding pixel coordinates in the orthophoto. Assuming the pixel resolution of the orthophoto is... (m / pixel, horizontal direction) and (m / pixel, vertical direction) represents the pixel coordinates of the intelligent inspection device on the orthophoto. It can be represented as:

[0039] in, These are the pixel coordinates of the detection starting point, manually selected on the orthophoto.

[0040] Furthermore, in this embodiment of the invention, the camera's shooting direction is calculated using the servo motor angle θ, and the covered area is marked with an arrow in the orthophoto. The starting point of the arrow is the robot's current position. The destination is:

[0041] in, This represents the pixel coordinates of the farthest point in the camera's field of view. Therefore, the coverage area and orientation of the image captured by the intelligent inspection device can be intuitively marked in the orthophoto, enabling the matching of location data and image data.

[0042] S4 divides the large-area continuous engineering facade orthophoto into horizontal and vertical grids and establishes a unique naming rule of "elevation - inspection unit - horizontal grid number - vertical grid number" to uniformly represent defect data.

[0043] To provide a clear and intuitive representation of damage to the surface of a large, continuous engineering facade, this invention divides the orthophoto of each individual inspection unit area into a grid along both the horizontal and vertical axes. Specifically, the horizontal axis divides the width of the inspection unit area into several equal parts, for example, six equal parts; the vertical axis is divided with meter-level accuracy based on the actual height of the dam surface.

[0044] In one embodiment of the invention, the height is divided into one grid per meter. Taking an inspection unit area with a height of 50 meters as an example, it is divided into 50 grid units vertically, and combined with 6 equal parts horizontally, a total of 6 × 50 = 300 grid areas are formed.

[0045] Furthermore, it should be noted that in the gridded orthophoto, the vertical lines of a preset color are used to represent the inspection trajectory of the intelligent inspection equipment. During each inspection, the intelligent inspection equipment starts from the bottom of the inspection unit area and climbs along this vertical trajectory to the top of the inspection unit area, realizing defect detection and data collection in the area covered by the trajectory.

[0046] In one possible embodiment, the yellow vertical lines are used to represent the inspection trajectory of the intelligent inspection device, such as... Figure 5 As shown.

[0047] Furthermore, through the cooperation of the winch drive motor and the encoder, this embodiment of the invention collects the cumulative number of encoder rotations. Given the physical length and width of the orthophoto of the inspection unit area, a detection starting point is first manually selected on the orthophoto. Then, the ascent or descent distance recorded by the encoder is converted into pixel dimensions to determine the robot's current position in the orthophoto. Further, based on the timestamp of the captured photo, time matching is performed with the encoder data, and the encoder record with the shortest time interval is selected to obtain the corresponding position of the photo captured by the intelligent inspection device in the orthophoto, and the position is marked on the image.

[0048] Based on this, embodiments of the present invention combine the rotation angle of the automatic scanning bracket servo motor of the intelligent inspection equipment with the field of view parameters of the camera to calculate the coverage area of ​​the image captured by the intelligent inspection equipment. Specifically, taking the image coordinates corresponding to the servo motor position as the starting point, the field of view direction and the farthest position are determined according to the servo motor angle and the length of the automatic scanning bracket, and a field of view line is drawn. The area covered by this field of view line is the effective field of view of the current image. If the image recognition result shows that there are cracks or other damage, all grid areas crossed by the field of view line are judged as "damaged areas" and filled and marked in the corresponding grid cells, thereby realizing the characterization of large-area continuous engineering facade defects based on gridding.

[0049] In one possible embodiment, a schematic diagram of the detected damage is shown below. Figure 6 As shown.

[0050] After completing the grid division of each inspection unit area of ​​the large-area continuous engineering facade, in order to accurately mark the location of each damage, this embodiment of the invention further establishes a grid coordinate system specifically for representing inspection data of intelligent inspection equipment. This coordinate system uses the generated orthophoto of the inspection unit area as a reference and uniquely names each grid unit according to the format of "elevation - inspection unit - horizontal grid number - vertical grid number". Its general form can be expressed as:

[0051] Where E represents the elevation number, B represents the inspection unit area number, i represents the horizontal grid number, and j represents the vertical grid number.

[0052] For example, in the inspection unit area B=23 with an elevation of E=1190, the grid cell in the horizontal column i=3 and vertical row j=5 can be named:

[0053] For ease of intuitive expression, this invention uses the string "EBij" for identification, i.e., "1190-23-3-5". This naming convention ensures that each grid cell has a unique number, enabling precise labeling and traceability of defect locations.

[0054] In one possible embodiment, a schematic diagram of the naming rules for the grid coordinate system of a large-area continuous engineering facade inspection is shown below. Figure 7 As shown.

[0055] S5: When the field of view of the image captured by the intelligent inspection device intersects with a certain grid cell, the number of images in that grid is incremented by 1, and the defect is marked in the grid, thereby completing the statistical analysis and visualization of large-area continuous engineering facade defect data based on grid cells.

[0056] Furthermore, since the field of view captured by the automatic scanning bracket of the intelligent inspection equipment often covers multiple grids simultaneously, and each grid may contain the coverage area of ​​multiple different images, it is necessary to count the number of images within each grid. The specific method is as follows: when the field of view of the intelligent inspection equipment... With a certain grid area When an intersection occurs, it is assumed that the grid contains a corresponding image, and its image count is incremented by 1. After a complete inspection, the... The number of images in a grid can be represented as:

[0057] in, This represents the number of images in the (i, j)th grid, where K is the total number of images collected during the inspection process. The field of view line corresponding to the k-th image. For the first Each grid area This is an indicator function; it takes the value 1 when the condition within the parentheses is true, and 0 otherwise. Using the above method, the total image coverage corresponding to each grid can be obtained and statistically displayed within the grid, thus achieving image coverage statistics and visualization based on grid cells.

[0058] In one possible embodiment, an example diagram of image data naming rules is shown below. Figure 8As shown.

[0059] In this embodiment of the invention, through the above steps, including the generation of orthophotos of the inspection unit area, the matching of positioning data and image data of the intelligent inspection equipment, the calculation and labeling of the field of view, the grid division and coordinate naming of the surface of a large-area continuous engineering facade, and the statistics and representation of the grid image data, a grid-based management method for defect data of large-area continuous engineering facades for intelligent inspection equipment applications is finally formed. This method can uniformly map the images and positioning information collected during the inspection process to the orthophoto coordinate system of a large-area continuous engineering facade, and achieve precise labeling and independent management of defect locations based on grid units. This constructs a unified representation system for defect data of large-area continuous engineering facades, significantly improving the traceability of inspection data and the efficiency of operation and maintenance management. In one possible embodiment, a schematic diagram of the practical application process from mesh generation to defect characterization is shown below. Figure 9 As shown.

[0060] Example 2 This invention also provides a large-area continuous engineering facade intelligent inspection grid positioning and data management device, which includes: The first module is used to acquire images of large-area continuous engineering facades, construct perspective transformation relationships using feature points of the inspection area of ​​large-area continuous engineering facades, perform geometric correction on the original images, and generate orthophotos with real physical size geometric proportions by combining the actual size of the inspection area of ​​large-area continuous engineering facades. The second module is used to calculate the climbing distance of the intelligent device along the suspension rope by using an encoder installed on the winch drive motor of the intelligent device and combining it with the diameter of the winch, and combine it with the starting point of the manually calibrated inspection unit area to map the position of the intelligent inspection device into the pixel coordinate system of the orthophoto. The third module is used to determine the shooting range of the camera based on the rotation angle of the automatic scanning bracket servo of the camera mounted on the intelligent inspection equipment, the length of the automatic scanning bracket, and the field of view parameters of the camera, and to mark the coverage area and orientation of the image captured by the intelligent inspection equipment in the orthophoto. The fourth module is used to divide the orthophotos of the large-area continuous engineering facade inspection area into horizontal and vertical grids, and to establish a unique naming rule of "elevation - inspection unit - horizontal grid number - vertical grid number" to uniformly represent the defect data. The fifth module is used to increment the image count of a grid cell by 1 when the field of view of the image captured by the intelligent inspection equipment intersects with a certain grid cell, and to mark the defect in the grid, thereby completing the statistical analysis and visualization of the appearance defects of a large-area continuous engineering facade based on grid cells.

[0061] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0062] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0063] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0064] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

Claims

1. A method for intelligent inspection grid positioning and data management of large-area continuous engineering facades, characterized in that, include: Acquire large-area continuous engineering facade images, construct perspective transformation relationships using feature points of the inspection work area shape, perform geometric correction on the original images, and generate orthophotos with true physical size geometric proportions by combining the actual dimensions of the large-area continuous engineering facades. By using the encoder of the winch drive motor installed on the intelligent inspection equipment, combined with the winch diameter, the climbing distance of the intelligent inspection equipment along the suspension rope is calculated, and combined with the start and end points of the large-area continuous engineering facade inspection operation area manually marked, the position of the intelligent inspection equipment is mapped to the pixel coordinate system of the orthophoto. Based on the angle of the automatic scanning bracket servo, the length of the automatic scanning bracket, and the field of view of the camera mounted on the intelligent inspection equipment, the shooting range of the camera is determined, and the coverage area and orientation of the image captured by the intelligent inspection equipment are marked in the orthophoto. The orthophotos of large-area continuous engineering facades are divided into horizontal and vertical grids, and a unique naming rule of "elevation - inspection unit - horizontal grid number - vertical grid number" is established to uniformly represent the defect data. When the field of view of the image captured by the intelligent inspection equipment intersects with a certain grid cell, the number of images in that grid is incremented by 1, and the defect is marked in the grid, thereby completing the statistical analysis and visualization of large-area continuous engineering facade defect data based on grid cells.

2. The method according to claim 1, characterized in that, The process of acquiring a large-area continuous engineering facade appearance image, constructing perspective transformation relationships using feature points of the inspection work area of ​​the large-area continuous engineering facade, performing geometric correction on the original image, and generating an orthophoto image with true physical size geometric proportions by combining the actual dimensions of the large-area continuous engineering facade includes: Images of large-area continuous engineering facades, including overall or partial inspection unit areas, are captured by drones, oblique photography, or cropped from 3D models or captured by fixed cameras. The actual height and length of each partial inspection unit area are known parameters. Select the coordinates of characteristic points in the inspection area of ​​a large-area continuous engineering facade. And arrange the vertices in order according to the requirements of perspective transformation; Constructing the perspective transformation matrix Geometric correction is performed on the original image to obtain an orthophoto; The perspective transformation relationship is expressed as follows: After expansion, we can obtain: in, These are the pixel coordinates in the original image. These are the pixel coordinates of the orthophoto after perspective transformation. These are the parameters of the perspective transformation matrix; After obtaining the orthophoto, a proportional mapping relationship between pixel coordinates and real physical coordinates is established on the orthophoto based on the known actual physical size of the inspection unit area.

3. The method according to claim 2, characterized in that, After obtaining the orthophoto image, based on the known actual physical dimensions of the inspection unit area, a proportional mapping relationship between the pixel coordinates and the actual physical coordinates of the orthophoto image is established, including: Let the pixel width of the inspection unit area in the orthophoto be... The pixel height is The actual physical length of the corresponding inspection unit area is The actual physical height is Then the pixel-to-physical-size scaling factors of the orthophoto in the horizontal and vertical directions are respectively: in, and These represent the actual physical dimensions of a unit pixel in the horizontal and vertical directions of an orthophoto image, respectively. For any pixel in an orthophoto and Its corresponding position in the actual physical coordinate system It can be represented as For the pixel distance between any two pixels in an orthophoto , The corresponding actual physical distance They are respectively This enables a quantitative mapping between the pixel scale in the orthophoto and the physical scale of the actual inspection unit area, so that the generated orthophoto simultaneously possesses geometric correction characteristics and the actual physical size ratio.

4. The method according to claim 3, characterized in that, By using the encoder of the winch drive motor installed on the intelligent inspection equipment and combining it with the winch diameter, the climbing distance of the intelligent inspection equipment along the suspension rope is calculated. This calculation is then combined with the manually calibrated starting point of the inspection unit area to map the position of the intelligent inspection equipment onto the pixel coordinate system of the orthophoto image, including: The climbing distance of the intelligent inspection equipment along the suspension rope direction is calculated using the following formula, expressed as: Where d is the climbing distance of the intelligent inspection equipment along the suspension rope, n is the number of motor rotations recorded by the encoder, and D is the diameter of the winch; Using the starting point calibration method, the initial position of the intelligent inspection equipment in the orthophoto of the inspection unit area is taken as the reference point. Based on the climbing distance d calculated by the encoder of the winch drive motor installed on the intelligent inspection equipment and the diameter of the winch, the position of the intelligent inspection equipment is mapped to the corresponding pixel coordinates of the orthophoto.

5. The method according to claim 4, characterized in that, The process of determining the camera's shooting range based on the angle of the automatic scanning bracket servo, the length of the automatic scanning bracket, and the camera's field of view on the intelligent inspection equipment, and marking the coverage area and orientation of the image captured by the intelligent inspection equipment in the orthophoto, includes: Based on the known servo angle θ, the length L of the automatic scanning bracket for mounting the camera on the intelligent inspection equipment, and the camera's field of view φ, the relative position of the coverage area of ​​the camera's image to the main body of the intelligent inspection equipment is determined. The coverage width range is expressed as: Where W is the coverage width of the camera in the inspection unit area; Assume the pixel resolution of the orthophoto is as follows in the horizontal and vertical directions: and The pixel coordinates of the intelligent inspection equipment on the orthophoto Represented as: in, The coordinates of the detection starting point pixel are manually selected on the orthophoto; The camera's shooting direction is calculated using the servo motor angle θ, and the coverage area is marked with an arrow on the orthophoto map, with the starting point of the arrow indicating the current position of the intelligent inspection device. The destination is: in, This indicates the pixel coordinates at the farthest point in the camera's field of view.

6. The method according to claim 5, characterized in that, The orthophoto image of the inspection unit area is divided into horizontal and vertical grids, and a unique naming rule of "elevation - inspection unit - horizontal grid number - vertical grid number" is established to uniformly represent the defect data, including: The orthophoto of each large-area continuous engineering facade inspection area is divided into grids in the horizontal and vertical directions. In the horizontal direction, the width of the inspection unit area is divided into several equal parts. In the vertical direction, the actual height of the large-area continuous engineering facade inspection area is divided with meter-level accuracy.

7. The method according to claim 6, characterized in that, Also includes: The trajectory of the intelligent inspection equipment is represented by a vertical line of preset color. During each inspection, the intelligent inspection equipment starts from the bottom of the large-area continuous engineering facade inspection work area and climbs to the top of the large-area continuous engineering facade inspection work area along the vertical trajectory to realize defect detection and data collection in the area covered by the trajectory. By cooperating with the winch drive motor and the encoder, the cumulative number of encoder rotations is collected. Given the physical length and width of the orthophoto of the inspection unit area, the detection starting point is manually selected on the orthophoto, and the upward or downward distance recorded by the encoder is converted into pixel size to determine the current position of the intelligent inspection device in the orthophoto. Based on the timestamp of the captured photo, time matching is performed with the encoder data, and the encoder record with the shortest time interval is selected to obtain the corresponding position of the photo captured by the intelligent inspection device in the orthophoto, and the position is marked on the image. By combining the rotation angle of the automatic scanning bracket servo of the camera mounted on the intelligent inspection equipment, the length of the automatic scanning bracket, and the field of view parameters of the camera, the coverage area of ​​the image captured by the intelligent inspection equipment is calculated.

8. The method according to claim 7, characterized in that, The establishment of a unique naming rule of "elevation - inspection unit - horizontal grid number - vertical grid number" is used to uniformly represent defect data, including: Each grid cell is uniquely named according to the format of "elevation - inspection unit - horizontal grid number - vertical grid number", and its general form is as follows: Where E represents the elevation number, B represents the inspection unit area number, i represents the horizontal grid number, and j represents the vertical grid number.

9. The method according to claim 8, characterized in that, When the field of view of an image captured by the intelligent inspection device intersects with a certain grid cell, the image count of that grid is incremented by 1, and the defect information is marked in the grid, including: When the field of vision of the intelligent inspection equipment With a certain grid area When an intersection occurs, it is assumed that the grid contains a corresponding image, and its image count is incremented by 1; After a complete inspection, the first The number of images in each grid is represented as: in, Indicates the first The number of images per grid, where K is the total number of images collected during the inspection process. The field of view line corresponding to the k-th image. For the first Each grid area This is an indicator function that takes the value 1 when the condition inside the parentheses is true, and 0 otherwise.

10. A large-area continuous engineering facade intelligent inspection grid positioning and data management device, characterized in that, include: The first module is used to acquire images of large-area continuous engineering facades, construct perspective transformation relationships using feature points of the inspection area of ​​large-area continuous engineering facades, perform geometric correction on the original images, and generate orthophotos with real physical size geometric proportions by combining the actual size of the inspection area of ​​large-area continuous engineering facades. The second module is used to calculate the climbing distance of the intelligent inspection equipment along the suspension rope by using an encoder installed on the winch drive motor of the intelligent equipment and combining it with the winch diameter, and to combine it with the starting point of the manually calibrated inspection unit area to map the position of the intelligent inspection equipment into the pixel coordinate system of the orthophoto. The third module is used to determine the shooting range of the camera based on the rotation angle of the automatic scanning bracket servo of the camera mounted on the intelligent inspection equipment, the length of the automatic scanning bracket, and the field of view parameters of the camera, and to mark the coverage area and orientation of the image captured by the intelligent inspection equipment in the orthophoto. The fourth module is used to divide the orthophotos of the large-area continuous engineering facade inspection area into horizontal and vertical grids, and to establish a unique naming rule of "elevation - inspection unit - horizontal grid number - vertical grid number" to uniformly represent the defect data. The fifth module is used to increment the image count of a grid cell by 1 when the field of view of the image captured by the intelligent inspection equipment intersects with a certain grid cell, and to mark the defect in the grid, thereby completing the statistical analysis and visualization of the appearance defects of a large-area continuous engineering facade based on grid cells.

Images