Method for identifying bumps of external wall tiles of external facade of building

Abstract

The invention discloses a building facade exterior wall brick bump identification method, and belongs to the technical field of building facade maintenance, the identification method depends on an unmanned aerial vehicle, the unmanned aerial vehicle is provided with a camera and a controller, and the controller is used for identifying an image shot by the camera; the identification method comprises the steps that the unmanned aerial vehicle is controlled to fly according to a preset path, the transverse shooting angle of a camera is adjusted so that the camera can face the exterior wall tiles of the building facade at an acute angle, and images of the exterior wall tiles are shot at a fixed visual angle regularly and sent to a controller; when the unmanned aerial vehicle is at a reference detection height, the controller constructs a theoretical two-dimensional grid template for the identified vertical gaps and transverse gaps; the problems that in existing building external wall brick bump detection, manual inspection visual angle deviation is caused, efficiency is low, danger is high, and bump recognition and severity evaluation are inaccurate due to the fact that three-dimensional features are difficult to highlight through front-view image shooting are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of building facade maintenance, and particularly to a method for identifying bulges on exterior wall tiles of a building facade. Background Art

[0002] In existing methods for detecting bulges on building exterior wall tiles, traditional manual detection requires operators to climb close or observe upward from the ground. This not only has low efficiency and high danger but is also limited by perspective deviation. When observing upward from the ground, due to the inclined line of sight, the exterior wall tiles undergo perspective distortion visually, making it difficult to accurately identify the actual position of the brick joints and the three-dimensional convex features of the bulging area. It is extremely easy to cause visual confusion between the edges of the bulges and normal wall tiles, resulting in missed detections, misjudgments, or misestimations of the severity of the bulges. For image recognition technologies based on frontal shooting, although attempts have been made to assist detection using drones, the differences in wall surface flatness under the frontal view are difficult to effectively present through two-dimensional images, and the shadow effects in the bulging areas are insufficient. There are still problems such as blurred edge features and difficulty in quantifying deviations. In addition, the establishment of theoretical grid templates in the existing technology lacks flexibility and is difficult to adapt to the arrangement rules of different building exterior wall tiles. Moreover, the detection path planning does not fully consider the detection efficiency and the stability of image acquisition, resulting in a longer detection process or larger image deviations. Summary of the Invention

[0003] The purpose of the present invention is to solve the above problems and provide a method for identifying bulges on exterior wall tiles of a building facade.

[0004] The technical solution of the present invention is implemented as follows: The present invention provides a method for identifying bulges on exterior wall tiles of a building facade. The identification method relies on a drone, and a camera and a controller are installed on the drone. The controller is used to identify the images captured by the camera; The identification method includes: controlling the drone to fly along a preset path and adjusting the horizontal shooting angle of the camera so that it faces the exterior wall tiles of the building facade at an acute angle, and regularly taking images of the exterior wall tiles at a fixed perspective and sending them to the controller; When the drone is at the reference detection height, the controller constructs a theoretical two-dimensional grid template for the identified vertical and horizontal joints. The theoretical two-dimensional grid template is a theoretical two-dimensional grid template generated based on the arrangement rules of standard exterior wall tiles; During subsequent flights, the controller compares the actual joint grid in the identified image with the theoretical grid template. If the deviation exceeds the threshold, the severity of the bulge is evaluated according to the degree of joint deviation; Wherein, the reference detection height is the height corresponding to the first or second floor of the building.

[0005] The advantages or beneficial effects in the above technical solutions at least include: 1. By controlling the drone to capture the images of exterior wall tiles at an inclined angle, the shadow effect of the bulging area is enhanced, the three-dimensional surface features are highlighted, the confusion between the edges of the bulge and normal wall tiles during frontal shooting is avoided, and the accuracy of bulge recognition is significantly improved.

[0006] 2. The theoretical two-dimensional grid template supports dual modes of automatic generation and manual selection. It can be automatically constructed based on the standard arrangement rules through image processing algorithms, or manually drawn by the staff in the normal area, improving the adaptability of the template to the arrangement characteristics of different exterior wall tiles of buildings.

[0007] 3. The preset path adopts a design combining vertical main flight segments and horizontal turning flight segments, covering the exterior facade mainly in the vertical direction, combined with optimized turning speed, shortening the turning time while ensuring the comprehensiveness of detection, and improving the overall detection efficiency.

[0008] 4. By comprehensively considering multiple factors such as the gap deviation value, area size, and deviation change gradient in the bulging area, the types and severity of the bulges are classified and evaluated, forming a more comprehensive and scientific basis for risk judgment.

[0009] 5. During the flight process, the coincidence degree between the theoretical grid and the actual gap is monitored in real time, and the template is automatically adjusted horizontally to match the actual gap, reducing the detection error caused by position offset or image distortion, and ensuring the stability of the detection result.

[0010] 6. After detecting a bulge, the position, height are automatically recorded and the image is uploaded to the terminal for manual review, realizing the organic combination of automated detection and manual intervention, and improving the convenience and reliability of the detection process. Description of the Drawings

[0011] The drawings illustrate exemplary embodiments of the present invention and, together with the description thereof, are used to explain the principles of the present invention. These drawings are included to provide a further understanding of the present invention and are included in this specification and form a part of this specification.

[0012] Figure 1 Shows a schematic diagram of the rotation angle of the drone camera in an embodiment of the present invention; Figure 2 Shows a schematic diagram of the image captured by the camera when the drone flies to the reference detection height in an embodiment of the present invention; Figure 3 Shows a schematic diagram of the controller generating a theoretical two-dimensional grid template from the image in an embodiment of the present invention; Figure 4 Shows a schematic diagram of the image captured by the camera when the drone flies to the detection area in an embodiment of the present invention, where the camera captures a bulge; Figure 5The schematic diagram shows the controller in the embodiment of the present invention generating an actual slit grid from an image; Figure 6 The schematic diagram shows the superposition and comparison of the actual slit grid with the theoretical two-dimensional grid template in the embodiment of the present invention; Figure 7 The schematic diagram shows the situation after the superposition and comparison in the embodiment of the present invention; Figure 8 The schematic diagram shows the preset flight path in the embodiment of the present invention; Figure 9 The schematic diagram shows the front view shooting in the embodiment of the present invention; Detailed implementation manners The embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although some embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. On the contrary, these embodiments are provided to more thoroughly and completely understand the present invention. It should be understood that the drawings and embodiments of the present invention are only for exemplary purposes and are not used to limit the protection scope of the present invention.

[0013] It should be noted that, without conflict, the implementation manners and features in the implementation manners of the present invention can be combined with each other. The present invention will be described in detail below with reference to the drawings and in combination with the implementation manners.

[0014] It should be understood that the term "including" and its variants used herein are open-ended, that is, "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description. It should be noted that the concepts such as "first" and "second" mentioned in the present invention are only used to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependent relationships.

[0015] It should be noted that the modifications of "one" and "multiple" mentioned in the present invention are illustrative rather than restrictive. Those skilled in the art should understand that unless otherwise clearly stated in the context, it should be understood as "one or more".

[0016] The names of the messages or information exchanged between multiple devices in the implementation manners of the present invention are only for illustrative purposes and are not used to limit the scope of these messages or information.

[0017] A method for identifying bulges in exterior wall tiles of a building facade, the identification method relying on a drone, on which a camera and a controller are installed, and the controller is used to identify the images captured by the camera; The identification method includes: controlling the drone to fly along a preset path, adjusting the horizontal shooting angle of the camera, regularly shooting images of the exterior wall tiles at a fixed perspective and sending them to the controller, so that it faces the exterior wall tiles of the building facade at an acute angle. When the drone is at the reference detection height, the controller constructs a theoretical two-dimensional grid template for the identified vertical and horizontal gaps. The theoretical two-dimensional grid template is a theoretical two-dimensional grid template generated based on the arrangement rules of standard exterior wall tiles; during subsequent flights, the controller compares the actual gap grid in the identified image with the theoretical grid template. If the deviation exceeds the threshold, the severity of the bulge is evaluated according to the degree of gap deviation; among them, the reference detection height is the height corresponding to the first or second floor of the building.

[0018] Specifically, as Figure 1 shown, the acute angle is from 30° to 60°, and preferably the acute angle is 45°. Figure 2 The enlarged part in Figure 4 is the image screen captured by the drone, including Figure 9 which is also the image screen captured by the drone. If taken from a frontal view, the edges of the wall tiles with protruding bulges in the screen are likely to be confused with the edges of the surrounding non-bulging wall tiles, making it difficult to judge the bulge and the severity of the bulge. Therefore, it is necessary to shoot at an inclined angle, as

[0019] shown, which can better highlight the three-dimensional features of the exterior wall tile surface and enhance the shadow effect of the bulge area.

[0019] Further, the preset path includes a vertical main flight segment and a horizontal turning flight segment, as Figure 8 shown; the vertical main flight segment extends along the vertical direction of the building facade, and the extension directions of adjacent vertical main flight segments are opposite; the horizontal turning flight segment connects the same-side endpoints of adjacent vertical main flight segments; this preset path can mainly check the wall tiles of the building facade in the vertical direction, and there is a fixed interval between the preset path and the building facade. Preferably, the fixed interval is 1.5 - 2m.

[0020] Among them, the flight speed of the vertical main flight segment is v1, and the flight speed of the horizontal turning flight segment is v2, 2*v2 ≥ v1, so as to shorten the time from turning from the vertical main flight segment to the adjacent vertical main flight segment.

[0021] In the control method, the part of the exterior wall tiles based on the theoretical two-dimensional grid template in the image can also be selected manually. The staff can control the drone to fly to a designated position outside the building, and the controller sends the image captured by the camera to the manual control terminal. The staff can control the drone through the control terminal to establish a theoretical two-dimensional grid template on the part of the normal exterior wall tiles without bulges, and even can hand-draw a theoretical two-dimensional template according to the arrangement of the exterior wall tiles at the reference detection height through the control terminal. As Figure 3 shown, the red line grid in the figure is the theoretical two-dimensional grid template.

[0022] Under the automatic generation of the theoretical two-dimensional network template, the controller identifies vertical and horizontal gaps in the following ways: preprocess the captured image, including grayscale conversion, filtering and denoising, and edge enhancement; apply the Canny edge detection algorithm to extract the edge information in the image. The Canny edge detection sets double thresholds (low threshold 30, high threshold 90) to extract edges and retain continuous and clear brick joint contours; identify straight edges through the Hough transform, and screen out vertical and horizontal gaps based on the arrangement characteristics of the exterior wall tiles. After identifying all vertical and horizontal gaps, calculate the spacing between each gap to obtain the size of the wall tiles in the image; construct a probability distribution model of the gap spacing through statistical analysis methods, and screen out the spacing value with the highest frequency of occurrence as the preliminary standard brick joint size; the controller identifies the complete cross-shaped brick joint intersection points in the preprocessed image as candidate origins; with the selected origin as the reference point, generate equally spaced grid lines along the horizontal and vertical directions according to the standard brick joint size. As Figure 3 shown, Figure 3 the red line grid in it is the theoretical two-dimensional grid template.

[0023] The ways for the controller to generate the actual gap grid include: Apply the probabilistic Hough transform to extract the edge information of the straight line segments in the image, and record the slope, starting coordinates and ending coordinates of each straight line segment; as Figures 4 to 5 shown. The generation method of the actual gap grid includes: extracting the actual vertical gap edge lines and horizontal gap edge lines in the image, and constructing to obtain the actual gap grid.

[0024] The method for the controller to compare the actually identified gap grid with the theoretical grid template includes: aligning the horizontal lines in the theoretical two-dimensional grid template with the horizontal and vertical lines of the actual gap grid, and the controller compares whether there is a gap between the vertical lines in the actual gap grid and the vertical lines in the theoretical two-dimensional template. As Figures 6 to 7As shown, by comparing the same position points in adjacent image frames, ensure that the origin positioning error is less than 0.5 times the standard brick joint size. If the gap exceeds the threshold, it is determined that a bulge has occurred. Prioritize selecting the intersection point in the central area of the image as the grid origin to reduce the influence of edge distortion on grid generation.

[0025] After the controller detects a bulge, it will evaluate the type and severity of the bulge. The severity is comprehensively evaluated based on the following factors: The maximum gap deviation value in the bulge area. The maximum gap deviation value in the bulge area is quantitatively evaluated by comparing the actually detected gap position with the maximum offset of the corresponding position in the theoretical grid template. The maximum gap deviation value is calculated based on the side length of the standard exterior wall brick as the reference unit; the maximum gap deviation value is divided into three grade intervals: when the deviation value is less than 0.1 times the side length of the standard brick, it is a mild bulge; when it is 0.1 - 0.3 times, it is a moderate bulge; when it is greater than 0.3 times, it is a severe bulge. The area size of the bulge area. The area size of the bulge area is evaluated by calculating the number of complete wall bricks contained in the closed area enclosed by continuously exceeding - standard gaps. The area size is measured in the number of standard exterior wall bricks; according to the exterior wall brick detachment risk assessment standard, the bulge area is divided into three grade intervals: when the area is less than 3 bricks, it is a mild bulge; when it is 3 - 10 bricks, it is a moderate bulge; when it is greater than 10 bricks, it is a severe bulge. The change gradient of the gap deviation value within the bulge area. The change gradient of the gap deviation value within the bulge area: The development trend of the bulge is evaluated by calculating the change rate of the gap deviation value between adjacent grid nodes. The change gradient is calculated based on the side length of the standard exterior wall brick as the reference unit; according to the analysis of the bulge morphological characteristics, the change gradient is divided into three grade intervals: when the gradient value is less than 0.05 times the side length of the brick, it is a gentle - type bulge; when it is 0.05 - 0.1 times the side length of the brick, it is a transitional - type bulge; when it is greater than 0.1 times the side length of the brick, it is a sharp - type bulge.

[0026] When establishing the above - mentioned theoretical two - dimensional grid template, the standard brick side - length unit of a single standard exterior wall brick has been obtained. By using the single standard brick side - length as the measurement unit for statistics, the error caused by the picture ratio and the real size can be reduced.

[0027] It should be noted that when the drone is flying along the preset flight path, the drone needs to adjust its position so that the vertical lines of the theoretical two - dimensional grid template in the image coincide with the vertical lines of the actual gap grid. Then fly up or down along the vertical main flight segment. Of course, during the flight process, it is also necessary to ensure the coincidence of the vertical lines. If a deviation occurs, the controller makes a horizontal adaptive adjustment to the theoretical two - dimensional grid template in the picture according to the actual gap grid until the two coincide.

[0028] When a bulge appears, the vertical lines of the actual gap grid can obviously deviate from the vertical lines of the theoretical two-dimensional grid template. As Figure 6 shown, after the controller detects the deviation, it stabilizes the height and attitude of the drone, stabilizes the picture, conducts identification, records the height and position, and then takes a photo and uploads it to the terminal for manual inspection, and then continues to conduct cruise identification.

[0029] In the description of the present invention, it should be noted that the orientation or positional relationship indicated by the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation to the present invention.

[0030] Those skilled in the art should understand that the above embodiments are only for clearly illustrating the present invention, rather than limiting the scope of the present invention. For those skilled in the art, other changes or modifications can be made on the basis of the above disclosure, and these changes or modifications are still within the scope of the present invention.

Claims

1. A method for identifying the bulging of exterior wall tiles on the building facade, characterized in that: The recognition method relies on a drone, on which a camera and a controller are installed, and the controller is used to recognize the images captured by the camera; The recognition method includes: controlling the drone to fly along a preset path, and adjusting the horizontal shooting angle of the camera so that it faces the exterior wall tiles of the building facade at an acute angle, and regularly shooting images of the exterior wall tiles at a fixed perspective and sending them to the controller; When the drone is at the reference detection height, the controller constructs a theoretical two-dimensional grid template for the identified vertical and horizontal gaps. The theoretical two-dimensional grid template is a theoretical two-dimensional grid template generated based on the arrangement rules of standard exterior wall tiles; During subsequent flights, the controller compares the actual gap grid in the recognized image with the theoretical grid template. If the deviation exceeds the threshold, the severity of the bulge is evaluated according to the degree of gap deviation; Among them, the reference detection height is the height corresponding to the first or second floor of the building.

2. The method for identifying the bulging of exterior wall tiles on the building facade according to claim 1, characterized in that: The controller identifies vertical and horizontal gaps in the following ways: Preprocess the captured images, including grayscale conversion, filtering and denoising, and edge enhancement; Apply the Canny edge detection algorithm to extract the edge information in the image; Identify straight edges through the Hough transform, and screen out vertical and horizontal gaps based on the arrangement characteristics of exterior wall tiles.

3. The method for identifying the bulging of exterior wall tiles on the building facade according to claim 2, characterized in that: The generation method of the theoretical two-dimensional grid template includes: After identifying all vertical and horizontal gaps, calculate the spacing between each gap; construct a probability distribution model of the gap spacing through statistical analysis methods, and screen out the spacing value with the highest frequency of occurrence as the preliminary standard brick joint size; The controller identifies the complete cross-shaped brick joint intersection in the preprocessed image as a candidate origin; taking the selected origin as the reference point, generate equally spaced grid lines along the horizontal and vertical directions according to the standard brick joint size; The way for the controller to generate the actual gap grid includes: Apply the probabilistic Hough transform to extract the edge information of the straight line segments in the image, and record the slope, starting coordinates and ending coordinates of each straight line segment; The generation method of the actual gap grid includes: Extract the actual vertical gap edge lines and horizontal gap edge lines in the image, and construct the actual gap grid; The method for the controller to compare the actually recognized gap grid with the theoretical grid template includes: Align the horizontal lines in the theoretical two-dimensional grid template with the horizontal and vertical lines of the actual gap grid. The controller compares whether there is a gap between the vertical lines in the actual gap grid and the vertical lines in the theoretical two-dimensional template. If the gap exceeds the threshold, it is judged that a bulge appears.

4. The method for identifying the bulging of exterior wall tiles on the building facade according to claim 1, wherein: The severity is comprehensively evaluated according to the following factors: The maximum gap deviation value of the bulge area. The maximum gap deviation value of the bulge area is quantitatively evaluated by comparing the maximum offset of the actually detected gap position with the corresponding position in the theoretical grid template. The maximum gap deviation value is calculated based on the side length of the standard exterior wall tile as the reference unit; the maximum gap deviation value is divided into three grade intervals: when the deviation value is less than 0.1 times the side length of the standard brick, it is a mild bulge, when it is 0.1 - 0.3 times, it is a moderate bulge, and when it is greater than 0.3 times, it is a severe bulge; The area size of the bulging area, and the area size of the bulging area is evaluated by calculating the number of complete wall tiles included in the closed area surrounded by continuously exceeding-standard gaps, and the area size is measured in the number of standard exterior wall tiles; according to the exterior wall tile falling-off risk assessment standard, the bulging area is divided into three grade intervals: when the area is less than 3 tiles, it is a mild bulge, when it is 3 - 10 tiles, it is a moderate bulge, and when it is more than 10 tiles, it is a severe bulge; The change gradient of the gap deviation value within the bulging area, and the change gradient of the gap deviation value within the bulging area: the development trend of the bulge is evaluated by calculating the change rate of the gap deviation value between adjacent grid nodes, and the change gradient is calculated based on the side length of the standard exterior wall tile as the reference unit; According to the analysis of the bulge morphological characteristics, the change gradient is divided into three grade intervals: when the gradient value is less than 0.05 times the side length of the tile, it is a gentle bulge, when it is 0.05 - 0.1 times the side length of the tile, it is a transitional bulge, and when it is more than 0.1 times the side length of the tile, it is a sharp bulge.

5. The method for identifying the bulging of exterior wall tiles on the building facade according to claim 1, wherein: The preset path includes: Vertical main flight segments, extending along the vertical direction of the building exterior, and the extension directions of adjacent vertical main flight segments are opposite; Horizontal turning flight segments, connecting the same-side endpoints of adjacent vertical main flight segments; Among them, the flight speed of the vertical main flight segment is v1, and the flight speed of the horizontal turning flight segment is v2, and 2*v2≥v1; Moreover, the detection ranges of the UAV in adjacent vertical main flight segments have an overlapping part; Moreover, there is a fixed interval between the preset path and the building exterior.

6. The method for identifying the bulging of exterior wall tiles on the building facade according to claim 1, characterized in that: The acute angle is 30° - 60°.

Images