A modeling method and application of ancient buildings based on UAV combined with close-range measurement

Abstract

The present invention discloses a modeling method and application of ancient buildings based on unmanned aerial vehicles combined with close-range measurement. The modeling method includes the following steps: formulating unmanned aerial vehicle aerial photography routes, setting up three-dimensional laser scanners and full-frame camera reshoot plans; selecting the positions of image control points and obtaining spatial coordinates; executing aerial photography, scanning tasks, and reshooting; generating dense point cloud data with an absolute coordinate system; preprocessing point cloud data; merging point cloud files preprocessed by unmanned aerial vehicles, close-range reshooting, and three-dimensional laser scanning to establish a real-life three-dimensional model of ancient buildings. Close-range photogrammetry with close-up reshooting enhances the texture expression of the model, and at the same time, it supplements complex structures and embedded structures that cannot be scanned by three-dimensional lasers and cannot be photographed by unmanned aerial vehicles. Compared with traditional contact measurement methods, the result data of the present invention has the advantages of being more intuitive, easier to save, higher precision, lower risk factor, convenient mapping, higher efficiency, etc., and the result data can be used in multiple ways.

Description

Technical Field

[0001] The invention belongs to the technical field of aerial photogrammetry, and in particular relates to a modeling method for ancient buildings based on unmanned aerial vehicles combined with close-range measurement. Background Art

[0002] Ancient buildings are precious cultural treasures left to us by our ancestors. They have historical, cultural, scientific, artistic and other values. A well-preserved ancient building is not only an important material for studying the history and culture of a certain period, but also a historical witness of social and cultural changes. As ancient cultural relics, ancient buildings have been subjected to the ruthless attacks of natural disasters such as wind, frost, rain, dew, water, fire, thunder and lightning day after day, year after year, which inevitably cause damage, destruction and collapse to ancient buildings. In addition, the situation of human destruction is becoming more and more serious, so the protection of ancient buildings is imperative.

[0003] As relics of ancient culture, ancient buildings are subjected to the ruthless attacks of natural disasters such as wind, frost, rain, dew, water, fire, lightning and thunder day after day and year after year, which inevitably cause damage, destruction and collapse of ancient buildings. Coupled with the increasingly serious situation of human destruction, the protection of ancient buildings is imperative.

[0004] From the perspective of surveying and mapping, according to the needs of different competent departments, participating in the protection of ancient buildings mainly depends on accurately, comprehensively and highly precisely obtaining the relevant geometric data and images of ancient buildings, including the general plan, plan, elevation and section drawings, main structural details, as well as orthophotos, three-dimensional models, disease images and descriptions, etc. These data can be widely used in the establishment and management of digital archives of ancient buildings, the renovation of ancient buildings, and even virtual tourism.

[0005] Most of the earliest data on ancient buildings were obtained by contact measurement. This method is not only inefficient but also has great safety hazards. It is not comprehensive and intuitive in terms of data, and is not convenient for archiving and management. Nowadays, most people use non-contact 3D scanning methods. Although it solves the problems of archiving and management and improves the measurement accuracy, it is difficult to obtain roof data for many single buildings. In addition, most ancient Chinese buildings are wooden structures with many complex carved structures such as brackets and eaves. Single 3D laser data can easily cause unclear patterns and textures in complex structural modeling models. It may also cause distortion of measurement data of structural components and key components. It is extremely difficult to obtain data on the roofs and attics of ancient buildings where it is inconvenient to set up 3D laser scanners, resulting in incomplete and intricate real-life 3D models, and there is not enough data support for later repairs and restorations of ancient buildings.

[0006] Therefore, a modeling method for ancient buildings based on UAVs combined with close-range measurement needs to be proposed urgently. Summary of the invention

[0007] In order to solve the defects of the prior art, the present invention provides a modeling method for ancient buildings based on unmanned aerial vehicles combined with close-range measurement.

[0008] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0009] The present invention provides a method for modeling ancient buildings based on unmanned aerial vehicles combined with close-range measurement, comprising the following steps:

[0010] S1. Conduct on-site survey of the operation area, formulate drone aerial photography routes, set up 3D laser scanners and plan for full-frame camera reshoots;

[0011] S2. Select the position of the image control points and obtain the spatial coordinates, and use the total station to measure the coordinates of the selected image control points;

[0012] S3. The drone takes aerial photos of the superstructure of the ancient building, the 3D laser scanner scans the inside and outside of the main body of the ancient building, and the full-frame camera takes supplementary photos of the details and key parts of the ancient building;

[0013] S4, performing aerial triangulation on the drone aerial photography data and the camera close-up photography data in combination with the image control point coordinates to generate dense point cloud data with an absolute coordinate system;

[0014] S5, preprocessing point cloud data of the three-dimensional laser scanning data;

[0015] S6. Combine the point cloud files pre-processed by drone, close-up supplementary shooting and 3D laser scanning to build a real-life 3D model of the ancient building.

[0016] Preferably, the aerial photography route in step S1 includes a group building route and a single building route, the group building route is carried out by a drone equipped with a gimbal camera to take pictures at equal time intervals or at equal distances in a manner perpendicular to the shooting subject; the single building route is carried out in a top-down circle with the geometric center of the single building as the orbiting point, the heading overlap of the aerial photography route is 53-60%, the lateral overlap is 15-30%, and the vertical overlap rate is 60%-80%.

[0017] Preferably, in step S3, the retaken photos are classified, and the photos taken in a dark place or against the light are processed for uniform light and color, so as to ensure that the modeling has sufficient texture features.

[0018] Preferably, the step S5 of preprocessing the point cloud data of the three-dimensional laser scanning data specifically comprises the following steps:

[0019] S51, data registration: Use the measured image control points to stitch the scanned data according to the image control points, and convert all point cloud data into the actual required coordinate system;

[0020] S52, data denoising: filtering the point cloud data using one of Gaussian filtering, mean filtering, and median filtering;

[0021] S53. Data simplification: Thin out data other than the scanned subject and with simple surface textures, and simplify the data to improve data operation speed, modeling rate and model accuracy.

[0022] Preferably, the present invention also provides an application of a real-life three-dimensional model of an ancient building, which is used to repair the ancient building by using a real-life three-dimensional model of the ancient building established by a modeling method of the ancient building based on drones combined with close-range measurement.

[0023] Preferably, the present invention also provides an application of a real-life three-dimensional model of an ancient building, which simulates natural disasters by using a real-life three-dimensional model of an ancient building established by a modeling method of ancient buildings based on drones combined with close-range measurement.

[0024] Compared with the prior art, the present invention has the following beneficial effects:

[0025] (1) The present invention uses a 3D laser scanner combined with drone oblique photography and close-range photogrammetry technology. 3D laser scanning makes up for the data accuracy of drone oblique photogrammetry, and drone oblique photogrammetry makes up for the defects of 3D laser top scanning blind spots and image acquisition viewing angles that are not wide enough, and also makes up for the problem that 3D lasers are difficult to set up on the top of ancient buildings. Close-range photogrammetry with close-up supplementary shooting enhances the texture performance of the model, and at the same time makes up for complex structures and embedded structures that cannot be scanned by 3D lasers or photographed by drones.

[0026] (2) Compared with the traditional contact measurement method, the result data of the present invention has the advantages of being more intuitive, easier to preserve, higher in accuracy, lower in risk factor, easier to draw, and more efficient, and the result data can be used in multiple ways.

[0027] (3) The present invention has high economic value for the promotion and research of ancient buildings. The real-life three-dimensional model can be combined with other technologies such as BIM reverse technology and geographic information platform to have an intuitive control over the health management and development trend of ancient buildings. In addition, the real-life three-dimensional model of ancient buildings can be linked on multiple platforms to simulate and analyze ancient buildings. DETAILED DESCRIPTION

[0028] The preferred embodiments of the present invention are described below. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0029] In this embodiment, the drone uses DJI MATRICE350RTK equipped with Zenmuse P1 integrated full-frame image sensor and three-axis gimbal. DJI MATRICE350RTK has a maximum flight time of 55 minutes, and a long single operation time; equipped with O3 image transmission and 4G enhanced image transmission to ensure safe flight in places where the signal is blocked or weak; equipped with Zenmuse P1 full-frame lens, 45 million pixel full-frame sensor, gimbal pitch angle of -130° to 40°, with better shooting quality and shooting angle.

[0030] In this embodiment, the 3D laser scanner is FARO Focus S330, which has a small size and a long scanning distance, is suitable for a variety of complex and harsh environmental conditions, has strong adaptability and fast operating speed.

[0031] In this embodiment, the image control point measurement adopts the Sokkia IM-101 high-precision measurement total station, which has a prism-free measurement mode, a prism-free measurement range of 1000 meters, a distance measurement accuracy of 1.5mm+2ppm, and an angle measurement accuracy of 1″, providing high precision for the coordinate measurement of the image control points.

[0032] In this embodiment, the full-frame camera is a Nikon Z7II full-frame camera with 45.75 million effective pixels.

[0033] This embodiment provides a method for modeling ancient buildings based on drone combined with close-range measurement, comprising the following steps:

[0034] S1. Conduct on-site survey of the operation area, formulate a plan for drone aerial photography routes, 3D laser scanner installation and full-frame camera reshoots.

[0035] For drone aerial photography routes: drones mainly shoot roof structures and components such as brackets, carvings, ridges, and pagodas. Aerial photography routes include group building routes and single building routes. The group building route is carried out by drones equipped with a gimbal camera to take photos at equal time intervals or distances in a manner perpendicular to the shooting subject; the single building route is carried out in a circular flight from top to bottom with the geometric center of the single building as the orbiting point. The heading overlap of the aerial photography route is 53-60%, the lateral overlap is 15-30%, and the vertical overlap rate is 60%-80%. The lateral overlap should be set according to the texture complexity of the area corresponding to the current shooting route. For tower buildings or buildings with brackets with more embedded structures, blind spots can be reduced by adding circular routes.

[0036] Most of the carvings or components are embedded under the brackets and are at a negative angle to the ground. In order to better obtain the texture information of the outer surface of the ancient buildings, the DJI MATRICE350RTK equipped with the Zenmuse P1 camera gimbal was set to shoot within the range of 10°-40°.

[0037] For the installation of 3D laser scanner: set the station position of 3D laser scanner, try to make the 3D laser scanner have line of sight with the scanning surface, set the target at the appropriate position, try to choose the place where there are few obstructions to ancient buildings or the texture of the ancient buildings is not rich; and the single station should be set up in a place where more data can be obtained as much as possible.

[0038] When scanning, first calibrate the 3D scanner, pre-set the scanning mode according to the 3D scanner, and calculate the position distance between the device and the ancient building. It is difficult to scan translucent or reflective materials (such as glass doors and windows and other reflective objects) and darker areas or components with weak light reflection. Spray a thin layer of developer on the corresponding position (the developer will not cause damage to the surface of the object or the human body, and can be washed off with clean water after the scanning is completed). The purpose is to scan the 3D features of the object later, and the data will be more accurate. However, if the developer is sprayed too much, the thickness of the object will be superimposed, which will affect the scanning accuracy. Therefore, the scanning accuracy needs to be verified after calibration. After confirming that the accuracy is correct, perform the scan according to the set measuring station.

[0039] For full-frame camera reshoots: reshoot blind spots of 3D scanners and places that require higher color reproduction. When reshooting, use an electric panoramic head to reshoot the reshoot area from top to bottom. The reshoot photos can also set an appropriate overlap rate based on actual conditions.

[0040] S2. Select the positions of image control points and obtain their spatial coordinates, and use a total station to measure the coordinates of the selected image control points.

[0041] The layout of image control points needs to follow the following basic principles: (1) The location of the image control points should be free of obvious obstructions and surrounded by relatively open areas; (2) The image control points should be located in locations that are easy to identify and have large texture differences; (3) The layout of image control points should be relatively uniform within the survey area to ensure that the entire survey area can be controlled. In locations where it is impossible to deploy image control points, additional control points should be added in the surrounding area to ensure modeling accuracy.

[0042] Specifically, the image control point measurement uses the Sokkia IM-101 high-precision measurement total station to set the three-dimensional coordinates of the measuring station. Set the coordinates of the backsight point or set the horizontal disk reading in the backsight direction as its azimuth. When the coordinates of the backsight point are set, the total station will automatically calculate the azimuth in the backsight direction and set the horizontal disk reading in the backsight direction as its azimuth. Set the prism constant to 0, and set the atmospheric correction value or the temperature and air pressure value. Measure the instrument height and prism height to 0 and enter them into the total station. Turn on the prism-free mode, aim at the target image control point and press the coordinate measurement key. The total station starts to measure the distance and calculate the three-dimensional coordinates of the displayed measuring point. Record the position and three-dimensional coordinate information of the image control points on site and make a point record book to facilitate the image control point puncture during point cloud registration and modeling and point cloud solution during aerial triangulation.

[0043] S3. The drone performs aerial photography of the superstructure of the ancient building, the 3D laser scanner performs scanning tasks on the inside and outside of the main body of the ancient building, and the full-frame camera takes supplementary photos of the details and key parts of the ancient building.

[0044] Since the on-site detailed reshooting will produce a large number of scattered photos, and the internal texture of the ancient buildings is highly similar, if the photos taken of the details are not reasonably classified, it will increase the difficulty and error rate when matching the image control points in the later aerial triangulation. The "photo compass classification method" is used to classify the reshot photos. The ancient building complex is represented by the building number of the single unit, which is divided into 1, 2, 3...n buildings. Each single ancient building is divided by layer from bottom to top, which is A, B, C...Z layer (generally the number of floors of ancient buildings will not exceed 26 floors), and the attic is represented by the floor number + A layer. Each floor is divided into areas according to the compass scale, with the north direction as 0, and one area is divided every 10 degrees clockwise, so a single floor is divided into 32 areas. If the area number of the photo of the southeast corner of the second floor of the No. 3 ancient building is expressed as 3B13. The image control points measured in this area use the same classification method to record the X, Y, and Z coordinates, and the area number is used as the point number to facilitate the puncture point operation of the image control points in the later aerial triangulation.

[0045] Some photos taken of details may be darker and have poor color saturation, such as those taken in a dark attic or against the light. Such photos need to be evenly distributed to ensure that the modeling has sufficient texture features. The evenly distributed method corrects the uneven distribution of brightness, contrast, hue, and saturation in one or more images, so that the brightness, contrast, hue, and saturation at each position of the image are basically consistent. The grayscale mean and variance of the local image can also be mapped to the given grayscale mean and variance values ​​through a filter. It is a local image transformation that makes the grayscale variance and grayscale mean at different positions of the image have approximately equal values, that is, the contrast of the area with small image contrast increases, and the contrast of the area with large image contrast decreases, so that the tiny information of grayscale in the image is enhanced.

[0046] S4. Perform aerial triangulation on the drone aerial photography data and the camera close-up photography data in combination with the image control point coordinates to generate dense point cloud data with an absolute coordinate system.

[0047] This embodiment adopts the beam method aerial triangulation. The beam method regional network aerial triangulation uses a beam of light composed of a picture as the basic unit of adjustment, and uses the collinear equation of the central projection as the basic equation of adjustment. Through the rotation and translation of each light beam in space, the light of the common point between the models is optimally intersected, and the image points of the heading and lateral direction are automatically matched by the image, and each flight network in the whole area is included in the coordinate system with a unified scale to form a loose regional network, and the approximate position of the external orientation elements and ground point coordinates of each picture is confirmed. The reference coordinate system corresponding to the image control point is selected and combined, and the camera parameters, POS files, GPS data and other related information are imported to create an aerial triangulation task. After the aerial triangulation is completed, the aerial triangulation tasks are copied respectively, and the control points are imported for point puncturing. After the point puncturing is completed, the adjustment is submitted. After the adjustment is completed, the aerial triangulation report and the accuracy of the control points of each aerial triangulation task need to be checked to confirm that there is no problem with each aerial triangulation task, thereby obtaining an encrypted point cloud file with absolute position information.

[0048] S5. Preprocess the point cloud data of the three-dimensional laser scanning data.

[0049] In the process of 3D laser scanning, the acquisition of point cloud data is often affected by factors such as object occlusion and uneven lighting, which can easily cause blind spots in the scanning of complex-shaped objects and form holes. At the same time, due to the limited scanning measurement range, large-sized objects or large-scale scenes cannot be fully measured at one time, and multiple scanning measurements are required. Therefore, the scanning results are often multiple blocks of point cloud data with different coordinate systems and noise, which cannot fully meet people's requirements for the authenticity and real-time performance of digital models. Therefore, it is necessary to perform denoising, simplification, registration and hole filling preprocessing on the 3D point cloud data. The preprocessing specifically includes the following steps:

[0050] S51, data registration: Use the measured image control points to stitch the scanned data according to the image control points, and convert all point cloud data into the actual required coordinate system;

[0051] S52, data denoising: filtering the point cloud data using one of Gaussian filtering, mean filtering, and median filtering;

[0052] S53. Data simplification: Thin out data other than the scanned subject and with simple surface textures, and simplify the data to improve data operation speed, modeling rate and model accuracy.

[0053] S6. The processed 3D laser scanning point cloud and the encrypted point cloud obtained by aerial triangulation are merged. To ensure the relative accuracy of the point cloud position, the merged point cloud is combined with the image control point coordinates for aerial triangulation, and then 3D reconstruction is performed to establish a real-life 3D model of the ancient building.

[0054] An application of a real-life three-dimensional model of an ancient building, which is used to repair the ancient building using a real-life three-dimensional model of the ancient building established by a modeling method of the ancient building based on unmanned aerial vehicles combined with close-range measurement.

[0055] With the passage of time, human factors and natural disasters, ancient buildings are suffering from varying degrees of damage all the time. The real-life 3D model of the ancient building can be combined with BIM to do model reverse engineering, and the ancient building model can be split into cross-section drawings, elevation drawings and other CAD drawings that are consistent with the actual size of the ancient building, which facilitates the repair of ancient buildings damaged by human factors and natural disasters.

[0056] An application of a real-life three-dimensional model of an ancient building, which simulates natural disasters by using a real-life three-dimensional model of an ancient building established by a modeling method of ancient buildings based on unmanned aerial vehicles combined with close-range measurement.

[0057] Although most of the damage to ancient buildings caused by natural disasters are force majeure, advance natural disaster simulation can greatly reduce the losses caused by natural disasters to ancient buildings.

[0058] Fire analysis: Most ancient buildings are made of wood, which is particularly important for fire prevention and control. The real-life 3D model of the ancient building can be imported into the fire simulation 3D software to mark the single building that needs to build an indoor 3D. Click to enter the interior of the building, create the internal structure, import the reference drawing and set the scale, adjust the overlap between the drawing and the model, and start building the floor and walls inside the building. After the construction is completed, add doors, windows, elevators, stairs and indoor fire-fighting facilities on the first floor. Return to the outdoors and add outdoor fire hydrants, water pump connectors and other outdoor fire-fighting facilities. Save the current scene. Click to enter the scene creation work, and you can familiarize yourself with key units, rehearse combat operations, or review combat evaluations in the built indoor 3D scene.

[0059] Earthquake analysis: Earthquake models can be built based on real-life 3D data, and the ecological space can be digitally mapped with real-life 3D data as the core, gathering professional data such as building lifeline projects, secondary disasters, shelters, and rescue forces. It serves the prevention and control of earthquake disaster risks and provides a scientific basis for the rapid disposal of earthquake disasters. On the basis of real-life 3D, highly realistic earthquake disaster simulation scenes can be achieved based on set earthquakes or earthquake simulations, and the shaking and destruction of buildings can be simulated in the digital twin simulation space to assess the disaster losses of buildings, lifeline systems, economy, and personnel. It is also possible to use the real-life 3D model of ancient buildings to build a scaled-down model of ancient buildings, and simulate the damage to the structure of ancient buildings caused by earthquakes on earthquake simulators.

[0060] Submergence analysis: The real-life 3D model of the ancient building can be imported into the WebGIS platform for 3D GIS submergence analysis. When used in rainy scenes, it can intuitively understand where water is likely to accumulate around the building, creating a risk of submergence. Corresponding anti-submergence measures can be taken based on local hydrological data.

[0061] In order to verify the accuracy of the results of the present invention, some characteristic points of a single building were selected and the Sokkia IM-101 high-precision measurement total station was used to measure the building walls, roofs and some characteristic points. The measurement results were compared with the final model results. The coordinate values ​​and coordinate differences are shown in Table 1.

[0062] It can be concluded from Table 1 that the maximum absolute value of the difference between the coordinates measured by the method of the present invention and the coordinates measured by the total station is 3 mm. It can be seen that the accuracy of the method of the present invention is comparable to that of the Sokkia IM-101 high-precision measurement total station, which meets the requirements of data accuracy in most aspects of ancient building protection. For example, in terms of ancient building repair, it can be close to the actual size of the original components.

[0063] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for modeling ancient buildings based on drone combined with close-range measurement, characterized in that: The following steps are involved: S1. Conduct on-site survey of the operation area, formulate drone aerial photography routes, set up 3D laser scanners and plan for full-frame camera reshoots; S2. Select the position of the image control points and obtain the spatial coordinates, and use the total station to measure the coordinates of the selected image control points; S3. The drone performs aerial photography on the superstructure of the ancient building, the 3D laser scanner performs scanning tasks on the inside and outside of the main body of the ancient building, and the full-frame camera performs supplementary photography on the details and key parts of the ancient building; the "photo compass classification method" is used to classify the supplementary photos, and the ancient building complex is represented by the building number of the individual building, which is divided into 1, 2, 3...n buildings. Each individual ancient building is divided by layers from bottom to top, that is, A, B, C...Z layers, and the attic is represented by the layer number + A layer; each layer is divided into regions according to the compass scale, with the north direction as 0, and one region is divided every 10 degrees clockwise, so a single layer is divided into 32 regions; the image control points measured in the area are recorded in the same classification method with X, Y, and Z coordinates, and the area number is used as the point number, which is convenient for the puncture point operation of the image control points during the later aerial triangulation; S4, respectively, the drone aerial data and the camera close-up shooting data are combined with the coordinates of the image control points to perform aerial triangulation to generate dense point cloud data with an absolute coordinate system; the beam method aerial triangulation is adopted, a beam of light composed of a picture is used as the basic unit of adjustment, and the collinear equation of the central projection is used as the basic equation of adjustment, and the light of the common point between the models is optimally intersected by rotating and translating each light beam in space, and the image points of the heading and lateral directions are automatically matched by the image, and each flight strip network in the whole area is incorporated into a coordinate system with a unified scale to form a loose regional network, and the approximate position of the exterior orientation elements and ground point coordinates of each picture is confirmed; the reference coordinate system corresponding to the image control points is selected and combined to create an aerial triangulation task, and after the aerial triangulation is completed, the aerial triangulation tasks are copied respectively, and the control points are imported for point puncturing. After the point puncturing is completed, the adjustment is submitted, and after the adjustment is completed, the aerial triangulation report of each aerial triangulation task and the accuracy of the control points are checked to confirm that there is no problem with each aerial triangulation task, thereby obtaining an encrypted point cloud file with absolute position information; S5, preprocessing point cloud data of the three-dimensional laser scanning data; S6. Combine the point cloud files pre-processed by drone, close-up supplementary shooting and 3D laser scanning to build a real-life 3D model of the ancient building.

2. The method for modeling ancient buildings based on drone combined with close-range measurement according to claim 1 is characterized in that: The aerial photography routes in step S1 include group building routes and single building routes. The group building routes are taken by a drone equipped with a gimbal camera at equal time intervals or distances in a manner perpendicular to the shooting subject; the single building routes are performed in a top-down circle with the geometric center of the single building as the orbiting point. The heading overlap of the aerial photography routes is 53-60%, the lateral overlap is 15-30%, and the vertical overlap rate is 60%-80%.

3. The method for modeling ancient buildings based on drone combined with close-range measurement according to claim 1 is characterized in that: In step S3, the supplementary photos are classified, and the photos taken in a dark place or against the light are processed for uniform light and color, so as to ensure that the modeling has sufficient texture features.

4. The method for modeling ancient buildings based on drone combined with close-range measurement according to claim 1 is characterized in that: The step S5 of preprocessing the point cloud data of the 3D laser scanning data specifically includes the following steps: S51, data registration: Use the measured image control points to stitch the scanned data according to the image control points, and convert all point cloud data into the actual required coordinate system; S52, data denoising: filtering the point cloud data using one of Gaussian filtering, mean filtering, and median filtering; S53. Data simplification: Thin out data other than the scanned subject and with simple surface textures, and simplify the data to improve data operation speed, modeling rate and model accuracy.

5. An application of a real-scene three-dimensional model of an ancient building, characterized in that: The ancient building is repaired by using a real-scene three-dimensional model of the ancient building established by the modeling method of the ancient building based on the unmanned aerial vehicle combined with close-range measurement as described in any one of claims 1 to 4.

6. An application of a real-scene three-dimensional model of an ancient building, characterized in that: A three-dimensional real-life model of an ancient building established by a modeling method for ancient buildings based on drones combined with close-range measurement as described in any one of claims 1 to 4 is used to simulate natural disasters.

Images