Crack detection and three-dimensional distribution system and method based on unmanned aerial vehicle and BIM technology

Abstract

According to the crack detection and three-dimensional distribution system and method based on the unmanned aerial vehicle and the BIM technology, data collection is conducted on a building outer wall through the unmanned aerial vehicle, an original picture of a crack on the surface of a building is obtained, a crack boundary is further extracted through a deep learning model, the width and length of the crack are calculated, and GPS data of the crack boundary are recorded; forming a crack parameter recording table; establishing a coordinate conversion relation between the real object wall and the model wall, and obtaining coordinates of the crack in the model wall through coordinate conversion; a crack parameterization family is established, three-dimensional distribution of the cracks in the BIM model is completed according to the width and length of the cracks in the crack parameter record table and the coordinates of the cracks in the model wall, the crack model can be established in the wall model, and therefore the positions and sizes of the cracks can be observed more visually; the problem of crack real mold consistency is solved, and the reliability of the building crack detection method is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of construction engineering, and in particular to a crack detection and three-dimensional distribution system and method based on unmanned aerial vehicle and BIM technology. Background Art

[0002] With the continuous advancement of modern urban construction, buildings such as high-rise residential buildings and office buildings occupy an important position in the city and become a beautiful landscape of urban development. However, over time, the exterior walls of these buildings may have various problems, such as cracks, falling off and other defects, which bring potential risks to public safety and the economy. Therefore, it is particularly important to conduct timely and efficient inspection and management of building exterior walls. In recent years, the rapid development of unmanned aerial vehicles (UAVs) has provided new possibilities for building exterior wall inspection. UAVs can be equipped with various sensors, such as infrared sensors, laser radars, etc. Through remote data collection technology, the risk of manual labor in dangerous environments is eliminated and the efficiency of data collection is improved.

[0003] Drone crack detection technology is already quite common, but this detection technology is often displayed in the form of pictures, and statistical analysis is performed on data such as the number of cracks and crack width. The crack results cannot be marked in the building model, and it is difficult for housing health assessors to evaluate the impact of cracks on the building.

[0004] Therefore, the existing building crack detection method has the technical problem of low reliability. It can be seen that how to improve the reliability of the building crack detection method is a problem that needs to be solved in this field. Summary of the invention

[0005] In view of the technical problem that the existing building crack detection methods have low reliability, the purpose of the present invention is to provide a crack detection and three-dimensional layout system and method based on drones and BIM technology. The cracks analyzed can be three-dimensionally laid out in the BIM model, and the position and size of the cracks can be observed more intuitively, which solves the problem of consistency between the actual model of the cracks and improves the reliability of the building crack detection method.

[0006] In order to achieve the above object, the present invention provides a method for crack detection and three-dimensional layout based on drone and BIM technology, and the method for crack detection and three-dimensional layout based on drone and BIM technology comprises the following steps:

[0007] Step 1: The drone is automatically controlled to maintain a fixed distance from the wall and fly to take photos. The drone collects data on the building's exterior wall, obtains original photos of cracks on the building's surface and forms a photo collection;

[0008] Step 2: Use the crack detection algorithm to identify crack targets on the image collected in step 1 and output an image with cracks;

[0009] Step 3: Then, the crack segmentation algorithm is further used to segment the cracked image outputted in step 3, and a crack mask map is outputted. The length and width of the crack are calculated, and then a proportional relationship between the photo pixels and the actual framing size is established, and the actual length and width of the crack are calculated and converted;

[0010] Step 4: Count the GPS coordinates, crack length, and crack width of each crack to form a crack parameter statistics table;

[0011] Step 5: Establish a coordinate transformation relationship between the physical wall and the model wall, and obtain the coordinates of the crack in the model wall through coordinate transformation;

[0012] Step 6: Create a parametric crack family and establish a parametric crack model. According to the crack width and length in the crack parameter record table, embed the crack model into the wall model to complete the three-dimensional layout of the crack in the BIM model.

[0013] Furthermore, in step 1, the drone includes a zoom camera, RTK positioning, and a lidar. The lidar module carried by the drone measures the vertical distance from the drone to the building surface in real time, so that the drone always maintains a fixed distance from the building facade during flight. The zoom camera photographs the wall during the flight of the drone and obtains the coordinates of the drone when the photo is taken through RTK positioning.

[0014] Furthermore, crack detection algorithm YOLOV8 is used to identify crack targets and output images with cracks.

[0015] Furthermore, the image with cracks output in step 2 is segmented by using the U-net crack segmentation algorithm to output a crack mask map.

[0016] Furthermore, the length between the start and end points of the crack mask is calculated and counted as the crack length; the width of the crack mask at multiple points perpendicular to the crack length is counted and the average value is calculated, and the average value is counted as the crack width.

[0017] Furthermore, a formula is used to establish a proportional relationship between the photo pixels and the actual framing size, and the photo pixel crack width and length are converted into the actual crack width and length. The specific formula is as follows:

[0018] The width and height of the field of view can be calculated by equation (1) and equation (2) respectively:

[0019]

[0020]

[0021] In the formula, F w is the field width, Fh is the height of the field of view, S w is the camera sensor width, S h is the camera sensor height, D is the vertical distance between the drone camera and the wall, and f is the focal length of the camera;

[0022] F w The field of view width is the actual viewing width, F h The field of view height is the actual viewing width. w The camera sensor width is the photo width, S h The camera sensor height is the photo height.

[0023] In order to achieve the above-mentioned object, the present invention provides a system for crack detection and three-dimensional layout based on UAV and BIM technology, which is characterized by comprising an acquisition module, a wireless transmission module, a crack identification and parameter extraction module, and a crack parameterized modeling module;

[0024] The acquisition module is used to collect data on the building's exterior wall to obtain original photos of cracks on the building's surface;

[0025] The wireless transmission module is arranged on the acquisition module, and is used to transmit the photos taken by the acquisition module to the crack identification and parameter extraction module in real time;

[0026] The crack identification and parameter extraction module is used to identify and output cracks in the photos collected by the acquisition module, calculate the parameters of the cracks, and extract them to the crack parameterization modeling module to establish a crack model in the wall model.

[0027] Furthermore, the acquisition module includes an acquisition unit, a positioning unit, and a distance detection unit;

[0028] The distance detection unit is used to measure the vertical distance between the acquisition unit and the building surface during the flight of the drone, and adjust the distance between the drone and the building facade according to the measured distance, so that the acquisition module always maintains a fixed distance from the building facade during the flight;

[0029] The acquisition unit takes photos of the building facade, and can obtain the coordinates of the drone when the photos are taken through the positioning unit.

[0030] Furthermore, the crack identification and parameter extraction module includes a crack identification unit, a crack segmentation unit, a parameter calculation unit, a parameter extraction unit and a parameter statistics unit;

[0031] The crack identification unit is used to identify cracks in the pictures collected by the collection module through a crack detection algorithm, and to screen the collected pictures and output images with cracks;

[0032] The crack segmentation unit is used to perform crack segmentation on the image with cracks output by the crack identification unit through a crack segmentation algorithm and output a crack mask map;

[0033] The parameter calculation unit is used to calculate the crack length and crack width of the crack mask image output by the crack segmentation unit, and then the parameter statistics unit performs statistics to form a crack parameter statistics table;

[0034] The parameter extraction unit is used to extract the data in the formed crack parameter statistical table to the crack parameter modeling module to establish a crack model in the wall model.

[0035] Furthermore, the crack parameterized modeling module includes a photo pixel acquisition unit, an actual framing size acquisition unit, a coordinate conversion unit and an import unit;

[0036] The actual framing size unit is used to obtain the width and height of the actual framing, and the pixel acquisition unit is used to obtain the width and height of the photo. The two use a formula to calculate the proportional relationship between the photo pixels and the actual framing size, and the actual width and length of the crack are obtained through calculation and conversion;

[0037] The coordinate conversion unit is used to establish a coordinate relationship between the wall entity and the wall photo, and convert the cracks identified in the wall photo into Revit coordinates through the GPS coordinates when the photo was taken;

[0038] The import unit is used to import the parameter statistics table counted by the parameter statistics unit, establish a parameterized crack model, and embed the established parameterized crack model into the wall model according to the transformed coordinates, thereby establishing a crack model in the wall model.

[0039] The crack detection and three-dimensional layout system and method based on drone and BIM technology provided by the present invention can calculate the collected crack image data to establish a parameterized crack model, and embed the established crack model into the wall model according to the conversion of each crack coordinate to form a crack model established in the wall model, thereby enabling a more intuitive observation of the position and size of the cracks, solving the problem of consistency of the actual model of the cracks, and improving the reliability of the building crack detection method. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments.

[0041] Figure 1 It is the technical route of UAV crack identification and model deployment in the present invention;

[0042] Figure 2 It is the path planning diagram of the UAV in the present invention;

[0043] Figure 3 This is a structural diagram of the drone data collection in the present invention;

[0044] Figure 4 A schematic diagram showing the relationship between the camera parameters and the field of view of the drone of the present invention;

[0045] Figure 5 It is a defect data mapping flow chart of the present invention;

[0046] Figure 6 It is a schematic diagram of the effect of three-dimensional distribution of cracks in the model of the present invention. DETAILED DESCRIPTION

[0047] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, the present invention is further explained below with reference to specific diagrams.

[0048] The method for detecting and three-dimensionally displaying building cracks based on drone and BIM technology provided by the present invention is described in detail in Figure 1 , which comprises the following steps:

[0049] Step 1: The drone is automatically controlled to maintain a fixed distance from the wall and fly to take photos. The drone collects data on the building's exterior wall, obtains original photos of cracks on the building's surface and forms a photo collection;

[0050] Step 2: Use the crack detection algorithm to identify crack targets on the image collected in step 1 and output an image with cracks;

[0051] Step 3: Then, the crack segmentation algorithm is further used to segment the cracked image output in step 3, output the crack mask map, and calculate the length and width of the crack. Then, the proportional relationship between the photo pixels and the actual framing size is established, and the actual length and width of the crack are calculated and converted.

[0052] Step 4: Count the GPS coordinates, crack length, and crack width of each crack to form a crack parameter statistics table.

[0053] Step 5: Establish a coordinate transformation relationship between the real wall and the model wall, and obtain the coordinates of the crack in the model wall through coordinate transformation.

[0054] Step 6: Create a parametric crack family and establish a parametric crack model. According to the crack width and length in the crack parameter record table, embed the crack model into the wall model to complete the three-dimensional layout of the crack in the BIM model.

[0055] Specifically, in step 1, in some embodiments, the drone includes equipment modules such as a zoom camera, RTK positioning, and a lidar, and flies parallel to the building facade. During the flight, the drone automatically adjusts the distance from the building surface.

[0056] For details, see Figure 2 , the laser radar module carried by the drone measures the vertical distance n from the drone to the building surface in real time. For example, according to the set vertical distance n1, when n is greater than or equal to n1+5cm, the drone moves 5cm toward the wall and then moves parallel to the original route; or when n is less than or equal to n1-5cm, the drone moves 5cm in the opposite direction of the wall and then moves parallel to the original route; through this method, the drone is prompted to always maintain a fixed distance from the building facade during flight; during flight, see Figure 3 , the drone’s zoom camera uses timing and viewing angles to take photos of building facades; RTK positioning can obtain the coordinates of the drone when the photo was taken.

[0057] Furthermore, the drone has a built-in wireless transmission module that can transmit the captured photos to the cloud in real time for crack identification and extraction.

[0058] Specifically, in step 2, in some embodiments, crack target recognition is performed using the crack detection algorithm YOLOV8, and an image with cracks is output.

[0059] Furthermore, in step 3, the image with cracks output in step 2 is segmented by using a U-net crack segmentation algorithm to output a crack mask map.

[0060] The specific steps of performing crack segmentation on the image with cracks output in step 2 by using the U-net crack segmentation algorithm and outputting the crack mask map are as follows:

[0061] 1. Data Preparation

[0062] Collect image data: Obtain a dataset of images containing cracks. These images should contain a variety of crack shapes, sizes, and background environments to ensure the generalization ability of the model.

[0063] Labeled data: In the collected image data, the crack areas in the data are labeled with specific colors, and the remaining areas are background colors.

[0064] 2. Model Training

[0065] Build a U-net model: Based on the U-net network architecture, use a deep learning framework (such as TensorFlow, PyTorch, etc.) to build a model.

[0066] Set the loss function and optimizer: Choose an appropriate loss function (such as cross entropy loss, Dice loss, etc.) and optimizer (such as Adam, SGD, etc.) to train the model.

[0067] Training model: Input the labeled image dataset into the U-net model for training. During the training process, the model will automatically learn the characteristics and segmentation rules of the cracks.

[0068] 3. Image Segmentation and Mask Generation

[0069] Preprocess the image: Preprocess the crack image to be segmented, such as denoising, contrast enhancement, etc., to improve the segmentation effect.

[0070] Input model for segmentation: Input the preprocessed image into the trained U-net model for segmentation. The model will output a segmentation result map of the same size as the input image.

[0071] Generate mask map: Extract the crack area according to the segmentation result map and generate the corresponding mask map. The mask map only contains the crack area, and the rest of the area is set as the background. Then calculate the length between the start and end of the crack mask and count it as the crack length. In the vertical crack length direction, count the width of the crack mask at multiple points, calculate the average value, and count the average value as the crack width.

[0072] Then the GPS coordinates, crack length and crack width of each crack are counted to form a crack parameter statistics table.

[0073] Here, the GPS coordinates of each crack are obtained by using the drone RTK positioning unit to obtain the coordinates of the drone when the photo was taken. The RTK positioning coordinates when the drone took the photo are the coordinates of the crack photo, and the crack must be in the center of the photo (or closest to the center of the photo). The method for judging whether the crack is in the center of the photo can be obtained by calculating the minimum value of the average pixel distance of each point of the crack relative to the center point of the photo, and the crack photo is obtained, and the RTK coordinates when the photo was taken are used as the center coordinates of the crack.

[0074] It should be noted here that if multiple photos of the same crack are taken, the data photos must be cleaned first. The crack photo is obtained by calculating the minimum value of the average pixel distance of each point of the crack relative to the center point of the photo, and the RTK coordinates when the photo was taken are used as the center coordinates of the crack.

[0075] Further, see Figure 4 , a formula is used to establish the proportional relationship between the photo pixels and the actual framing size, and the photo pixel crack width and length are converted into the actual crack width and length. The specific formula is as follows:

[0076] The width and height of the field of view can be calculated by equation (1) and equation (2) respectively.

[0077]

[0078]

[0079] In the formula, F w is the field width, F h is the height of the field of view, S w is the camera sensor width, S h is the camera sensor height, D is the vertical distance between the drone camera and the wall, and f is the focal length of the camera.

[0080] F w The field of view width is the actual viewing width, F h The field of view height is the actual viewing width. w The camera sensor width is the photo width, S h The camera sensor height is the photo height.

[0081] The proportional relationship between the photo pixel and the actual framing size is established through formula (1) and formula (2), and the actual distance size corresponding to a single pixel of the photo is obtained. The calculated crack width and length of the photo pixel are combined with the actual distance size corresponding to a single pixel, and the actual width and length of the crack are obtained through calculation and conversion.

[0082] The coordinate conversion relationship is established in step 5. First, use the drone to take photos at the lower left corner and the upper right corner of the wall respectively. RTK positioning can obtain the coordinates of the drone when the photos were taken, which are (N1x, N1y, N1z) and (N2x, N2y, N2z).

[0083] The coordinates of the corresponding lower left corner and upper right corner of the model wall are (M1x, M1y, M1z) and (M2x, M2y, M2z) respectively.

[0084] Given reference points M1, M2 (revit coordinates), N1, N2 (GPS coordinates), and M1 corresponds to N1, M2 corresponds to N2. Calculate the scaling factor and offset using the following formula:

[0085]

[0086]

[0087]

[0088] Among them, M1 X ,M1 Y ,M1 Z and M2 X ,M2 Y ,M2 Z are the coordinates of two reference points in Revit coordinates; N1 X ,N1Y ,N1 Z and N2 X ,N2 Y ,N2 Z are the coordinates of two reference points in GPS coordinates.

[0089] offset X =M1 X -N1 X ×scale_factor X

[0090] offset Y =M1 Y -N1 Y ×scale_factor Y

[0091] offset Z =M1 Z -N1 Z ×scale_factor Z

[0092] Then perform coordinate transformation

[0093] M X =X GPS ×scale_factor X +offset X

[0094] M Y =Y GPS ×scale_factor Y +offset Y

[0095] M Z =Z GPS ×scale_factor Z +offset Z

[0096] Among them, X GPS ,Y GPS ,Z GPS is the new GPS coordinate; the converted M X ,M Y ,M Z are the corresponding Revit coordinates.

[0097] By establishing a coordinate relationship between the wall entity and the wall photo, the cracks identified in the wall photo are converted into Revit coordinates using the GPS coordinates when the photo was taken.

[0098] Then, through the Add-in manager (Manual Mode) module, the statistical table of crack parameters is imported, and the crack model is embedded into the wall model according to the crack width and length in the crack parameter record table. Figure 6 , complete the three-dimensional layout of cracks in the BIM model. Based on the above-mentioned solution, the building crack detection and three-dimensional layout method based on drone and BIM technology, this solution also provides a building crack detection and three-dimensional layout system based on drone and BIM technology, which can realize the above-mentioned building crack detection and three-dimensional layout method based on drone and BIM technology.

[0099] The building crack detection and three-dimensional layout system based on UAV and BIM technology includes an acquisition module, a wireless transmission module, a crack identification and parameter extraction module, and a crack parameterized modeling module.

[0100] The acquisition module is used to collect data on the building's exterior wall to obtain original photos of cracks on the building's surface. The acquisition module includes an acquisition unit, a positioning unit, and a distance detection unit.

[0101] The distance detection unit is used to measure the vertical distance from the acquisition unit to the building surface during the flight of the drone. The distance between the drone and the building facade is adjusted according to the measured distance, so that the acquisition module always maintains a fixed distance from the building facade during the flight.

[0102] During the flight, the collection unit takes photos of the building facade, and the positioning unit can obtain the coordinates of the drone when the photo was taken.

[0103] The acquisition module in this solution preferably uses a drone, which can quickly and accurately collect original photos of cracks on the surface of the building. The drone includes a zoom camera, RTK positioning, and a lidar device, wherein the lidar is used to measure the vertical distance from the drone to the building surface during the flight of the drone. The distance between the drone and the building facade is adjusted according to the measured distance, so that the drone always maintains a fixed distance from the building facade during the flight.

[0104] During the flight, photos of the building facades are taken through the zoom camera, and the coordinates of the drone when the photos were taken can be obtained through RTK positioning.

[0105] Furthermore, a wireless transmission module is provided on the acquisition module, and is used to transmit the photos taken by the acquisition module to the crack identification and parameter extraction module in real time.

[0106] In some embodiments, the wireless transmission module preferably has a built-in 5G module, which has a fast transmission speed and a good signal, and can quickly and completely transmit the captured photos to the crack identification and parameter extraction module for crack identification and parameter extraction.

[0107] The crack identification and parameter extraction module is used to identify and output cracks in the photos collected by the acquisition module and calculate and extract crack parameters. The crack identification and parameter extraction module includes a crack identification unit, a crack segmentation unit, a parameter calculation unit, a parameter extraction unit and a parameter statistics unit.

[0108] The crack recognition unit is used to identify cracks in the pictures collected by the collection module through a crack detection algorithm, and to screen the collected pictures and output images with cracks.

[0109] The crack segmentation unit is used to perform crack segmentation on the image with cracks output by the crack identification unit through a crack segmentation algorithm and output a crack mask map.

[0110] The parameter calculation unit is used to calculate the length between the start and end points of the crack mask image output by the crack segmentation unit, which is counted as the crack length. In the vertical crack length direction, the width of the crack mask at multiple points is counted, and the average value is calculated, which is counted as the crack width.

[0111] Then, the parameter statistics unit is used to count the GPS coordinates, crack length, and crack width of each crack to form a crack parameter statistics table.

[0112] The parameter extraction unit is used to extract the data in the formed crack parameter statistical table to the crack parameter modeling module to establish a crack model in the wall model.

[0113] The crack parameterized modeling module includes a photo pixel acquisition unit, an actual framing size acquisition unit, a coordinate conversion unit and an import unit.

[0114] The actual framing size unit is used to obtain the width and height of the actual framing, and the pixel acquisition unit is used to obtain the width and height of the photo. Then the two are used to calculate the ratio of the photo pixel to the actual framing size to obtain the actual distance size corresponding to a single pixel of the photo. The calculated photo pixel crack width and length are combined with the actual distance size corresponding to a single pixel, and the actual width and length of the crack are obtained through calculation and conversion.

[0115] The coordinate conversion unit is used to establish the coordinate relationship between the wall entity and the wall photo, and convert the cracks identified in the wall photo into Revit coordinates through the GPS coordinates when the photo was taken.

[0116] See also Figure 5-Figure 6The import unit is used to import the parameter statistics table counted by the parameter statistics unit, establish a parametric crack model, and embed the established parametric crack model into the wall model according to the transformed coordinates, thereby completing the three-dimensional layout of the cracks in the BIM model.

[0117] The building crack detection and three-dimensional layout system and method based on drone and BIM technology composed of the above scheme can calculate the collected crack image data to establish a parameterized crack model, and embed the established crack model into the wall model according to the conversion of each crack coordinate to form a crack model established in the wall model, thereby enabling a more intuitive observation of the location and size of the crack, solving the problem of consistency of the crack real model, and improving the reliability of the building crack detection method. It can be applied to structural detection and evaluation, historical building protection, construction and renovation, disaster assessment and recovery, architectural design verification and other fields.

[0118] The above shows and describes the basic principles, main features and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for crack detection and three-dimensional layout based on drone and BIM technology, characterized in that: The method for crack detection and three-dimensional layout based on drone and BIM technology includes the following steps: Step 1: The drone is automatically controlled to maintain a fixed distance from the wall and fly to take photos. The drone collects data on the building's exterior wall, obtains original photos of cracks on the building's surface and forms a photo collection; Step 2: Use the crack detection algorithm to identify crack targets on the image collected in step 1 and output an image with cracks; Step 3: Then, the crack segmentation algorithm is further used to segment the cracked image outputted in step 3, and a crack mask map is outputted. The length and width of the crack are calculated, and then a proportional relationship between the photo pixels and the actual framing size is established, and the actual length and width of the crack are calculated and converted; Step 4: Count the GPS coordinates, crack length, and crack width of each crack to form a crack parameter statistics table; Step 5: Establish a coordinate transformation relationship between the physical wall and the model wall, and obtain the coordinates of the crack in the model wall through coordinate transformation; Step 6: Create a parametric crack family and establish a parametric crack model. According to the crack width and length in the crack parameter record table, embed the crack model into the wall model to complete the three-dimensional layout of the crack in the BIM model.

2. The method for crack detection and three-dimensional layout based on drone and BIM technology according to claim 1 is characterized in that: In step 1, the drone includes a zoom camera, RTK positioning, and a laser radar. The laser radar module carried by the drone measures the vertical distance from the drone to the building surface in real time, so that the drone always maintains a fixed distance from the building facade during flight. The zoom camera takes pictures of the wall during the flight of the drone and obtains the coordinates of the drone when the picture is taken through RTK positioning.

3. The method for crack detection and three-dimensional layout based on drone and BIM technology according to claim 1 is characterized in that: The crack detection algorithm YOLOV8 is used to identify crack targets and output images with cracks.

4. The method for crack detection and three-dimensional layout based on drone and BIM technology according to claim 1 is characterized in that: The image with cracks output in step 2 is segmented by the U-net crack segmentation algorithm to output a crack mask map.

5. The method for crack detection and three-dimensional layout based on drone and BIM technology according to claim 4 is characterized in that: The length between the start and end points of the crack mask is calculated as the crack length; the width of the crack mask at multiple points perpendicular to the crack length is counted, the average value is calculated, and the average value is counted as the crack width.

6. The method for crack detection and three-dimensional layout based on drone and BIM technology according to claim 1 is characterized in that: The proportional relationship between the photo pixels and the actual framing size is established through a formula, and the width and length of the photo pixel crack are converted into the width and length of the actual crack. The specific formula is as follows: The width and height of the field of view can be calculated by equation (1) and equation (2) respectively: In the formula, F w is the field width, F h is the height of the field of view, S w is the camera sensor width, S h is the camera sensor height, D is the vertical distance between the drone camera and the wall, and f is the focal length of the camera; F w The field of view width is the actual viewing width, F h The field of view height is the actual viewing width. w The camera sensor width is the photo width, S h The camera sensor height is the photo height.

7. A crack detection and three-dimensional layout system based on drone and BIM technology, characterized in that: It includes acquisition module, wireless transmission module, crack identification and parameter extraction module, and crack parameter modeling module; The acquisition module is used to collect data on the building's exterior wall to obtain original photos of cracks on the building's surface; The wireless transmission module is arranged on the acquisition module, and is used to transmit the photos taken by the acquisition module to the crack identification and parameter extraction module in real time; The crack identification and parameter extraction module is used to identify and output cracks in the photos collected by the acquisition module, calculate the parameters of the cracks, and extract them to the crack parameterization modeling module to establish a crack model in the wall model.

8. The system for crack detection and three-dimensional layout based on drone and BIM technology according to claim 7 is characterized in that: The acquisition module includes an acquisition unit, a positioning unit, and a distance detection unit; The distance detection unit is used to measure the vertical distance between the acquisition unit and the building surface during the flight of the drone, and adjust the distance between the drone and the building facade according to the measured distance, so that the acquisition module always maintains a fixed distance from the building facade during the flight; The acquisition unit takes photos of the building facade, and can obtain the coordinates of the drone when the photos are taken through the positioning unit.

9. The system for crack detection and three-dimensional layout based on drone and BIM technology according to claim 7, characterized in that: The crack identification and parameter extraction module includes a crack identification unit, a crack segmentation unit, a parameter calculation unit, a parameter extraction unit and a parameter statistics unit; The crack identification unit is used to identify cracks in the pictures collected by the collection module through a crack detection algorithm, and to screen the collected pictures and output images with cracks; The crack segmentation unit is used to perform crack segmentation on the image with cracks output by the crack identification unit through a crack segmentation algorithm and output a crack mask map; The parameter calculation unit is used to calculate the crack length and crack width of the crack mask image output by the crack segmentation unit, and then the parameter statistics unit performs statistics to form a crack parameter statistics table; The parameter extraction unit is used to extract the data in the formed crack parameter statistical table to the crack parameter modeling module to establish a crack model in the wall model.

10. The system for crack detection and three-dimensional layout based on drone and BIM technology according to claim 7, characterized in that: The crack parameterized modeling module includes a photo pixel acquisition unit, an actual framing size acquisition unit, a coordinate conversion unit and an import unit; The actual framing size unit is used to obtain the width and height of the actual framing, and the pixel acquisition unit is used to obtain the width and height of the photo. The two use a formula to calculate the proportional relationship between the photo pixels and the actual framing size, and the actual width and length of the crack are obtained through calculation and conversion; The coordinate conversion unit is used to establish a coordinate relationship between the wall entity and the wall photo, and convert the cracks identified in the wall photo into Revit coordinates through the GPS coordinates when the photo was taken; The import unit is used to import the parameter statistics table counted by the parameter statistics unit, establish a parameterized crack model, and embed the established parameterized crack model into the wall model according to the transformed coordinates, thereby establishing a crack model in the wall model.

Images