阅读说明：本技术 一种基于无人机摄像的震损结构层间残余变形检测方法 (Seismic damage structure interlayer residual deformation detection method based on unmanned aerial vehicle camera shooting ) 是由 侯爽 陈华儒 吴承治 张钜键 符明东 陈麟君 于 2021-05-08 设计创作，主要内容包括：本发明提供了一种基于无人机摄像的震损结构层间残余变形检测方法。所述方法包括如下步骤：使无人机按照设定的飞行路线进行飞行并对建筑进行全面拍摄；将拍摄得到的照片传送到地面控制平台；所述地面控制平台接收照片后,利用图像识别技术及三维重构技术对拍摄的建筑进行三维重构,计算建筑的层间位移角,完成震损结构层间残余变形检测。本发明可以突破地形的限制,实现全方位、多角度检测；同时,图像识别可以降低人为误判的概率；再者,采用无人机进行拍照方便、快捷,可大大减少检测时间。(The invention provides a seismic damage structural interlayer residual deformation detection method based on unmanned aerial vehicle camera shooting. The method comprises the following steps: the unmanned aerial vehicle flies according to the set flying route and takes a full shot of the building; transmitting the shot picture to a ground control platform; and after receiving the picture, the ground control platform carries out three-dimensional reconstruction on the shot building by utilizing an image recognition technology and a three-dimensional reconstruction technology, calculates the interlayer displacement angle of the building and finishes the interlayer residual deformation detection of the seismic damage structure. The invention can break through the limitation of the terrain and realize omnibearing and multi-angle detection; meanwhile, the probability of artificial misjudgment can be reduced by image identification; moreover, adopt unmanned aerial vehicle to shoot convenient, swift, detection time that can significantly reduce.)

1. The utility model provides a seismic damage structure interlaminar residual deformation detection method based on unmanned aerial vehicle makes a video recording, which is characterized in that, includes the following steps:

s1, enabling the unmanned aerial vehicle to fly according to the set flight route and shoot the building comprehensively;

s2, transmitting the shot picture to a ground control platform;

and S3, after receiving the picture, the ground control platform carries out three-dimensional reconstruction on the shot building by using an image recognition technology and a three-dimensional reconstruction technology, calculates the interlayer displacement angle of the building and finishes the interlayer residual deformation detection of the seismic damage structure.

2. The method for detecting the residual deformation between the earthquake damage structure layers based on the unmanned aerial vehicle camera shooting is characterized in that in the step S1, the flight routes comprise 4 transverse arc-shaped flight routes, and the route conversion operation needs to be carried out for 3 times; the starting position and the end position of each transverse arc-shaped flight path are the same; the arc flying routes are distributed at equal intervals from bottom to top.

3. The method for detecting the residual deformation between earthquake damage structural layers based on unmanned aerial vehicle camera shooting is characterized in that the flight speed and the shooting time interval are fixed at the initial position of each transverse arc-shaped flight path until the transverse arc-shaped flight path is completed; and controlling the unmanned aerial vehicle to move to the initial position of the transverse arc flight path of the previous layer at the end position of each transverse arc flight path, and repeating the previous step until the shooting of all the flight paths is completed.

4. The method for detecting the residual deformation between earthquake damage structural layers based on the camera shooting of the unmanned aerial vehicle as claimed in claim 1, wherein the flying speed of the unmanned aerial vehicle is fixed during the flying process of the unmanned aerial vehicle, and the shooting time interval is set according to the building size and flying speed required to be shot.

5. The method for detecting the residual deformation between earthquake damage structural layers based on the unmanned aerial vehicle camera shooting as claimed in claim 1, wherein in the flight process of the unmanned aerial vehicle, the camera lens always faces to the front of the unmanned aerial vehicle and vertically inclines downwards by 45 degrees.

6. The method for detecting the residual deformation between the earthquake damage structure layers based on the unmanned aerial vehicle camera shooting is characterized in that in the step S3, all photos shot in the flying process are subjected to image preprocessing, feature point monitoring and matching and dense point cloud generation by using Photoscan; establishing a right-hand coordinate system by taking the center of the bottom of the building as an origin and the inclination direction of the building as the positive direction of an X axis to finish three-dimensional reconstruction; and reading and processing point cloud data through Python, and solving an interlayer displacement angle.

7. The method for detecting the residual deformation between earthquake damage structural layers based on unmanned aerial vehicle camera shooting as claimed in claim 6, wherein the three-dimensional reconstruction mode is as follows:

and (5) performing three-dimensional reconstruction of the seismic damage structure on the screened photos by using Photoscan, and deriving point cloud data with three-dimensional coordinate axes.

8. The method for detecting the residual deformation between earthquake damage structural layers based on unmanned aerial vehicle camera shooting is characterized in that when the interlayer displacement angle of the building structure is detected to be larger than a specified value, the detection position is recorded and an alarm is given.

Technical Field

The invention relates to the field of earthquake damage building structure detection, in particular to an unmanned aerial vehicle camera shooting-based earthquake damage structure interlayer residual deformation detection method.

Background

China is a country with multiple earthquakes, and a large number of buildings are damaged when the earthquake occurs. The earthquake damage evaluation method has the advantages that accurate and rapid earthquake damage identification and safety evaluation can be carried out on earthquake damage buildings after earthquake, and the earthquake damage evaluation method has important significance for earthquake damage evaluation, post-disaster rescue and reconstruction and the like. At present, the earthquake damage assessment work of buildings on the earthquake field in China is mainly carried out on-site detection by professional staff by means of instruments and equipment. Currently, this detection method has the following limitations: the assessment personnel are limited, and the workload is large; the traditional damage detection method needs to be carried out by means of a precise instrument, and is not suitable for the evaluation work of a large number of buildings after a large amount of earthquakes; the complex and serious topography in the earthquake-stricken area affects the development of evaluation work, and for buildings with serious earthquake damage, the buildings are not suitable to be approached due to safety consideration, and enough earthquake damage information is difficult to obtain.

Based on the difficulties, the invention provides an earthquake damage structure interlayer residual deformation detection method based on unmanned aerial vehicle camera shooting.

Disclosure of Invention

The invention aims to provide a method for detecting residual deformation between earthquake damage structural layers by using an unmanned aerial vehicle based on an image recognition technology, so as to save labor and ensure reliable results.

The purpose of the invention is realized by at least one of the following technical solutions.

A seismic damage structure interlayer residual deformation detection method based on unmanned aerial vehicle shooting comprises the following steps:

s1, enabling the unmanned aerial vehicle to fly according to the set flight route and shoot the building comprehensively;

s2, transmitting the shot picture to a ground control platform;

and S3, after receiving the picture, the ground control platform carries out three-dimensional reconstruction on the shot building by using an image recognition technology and a three-dimensional reconstruction technology, calculates the interlayer displacement angle of the building and finishes the interlayer residual deformation detection of the seismic damage structure.

Further, in step S1, the flight path includes 4 transverse arc flight paths, and 3 times of route switching operations are required; the starting position and the end position of each transverse arc-shaped flight path are the same; the arc flying routes are distributed at equal intervals from bottom to top.

Further, at the initial position of each transverse arc-shaped flight path, the flight speed and the shooting time interval are fixed until the transverse arc-shaped flight path is completed; and controlling the unmanned aerial vehicle to move to the initial position of the transverse arc flight path of the previous layer at the end position of each transverse arc flight path, and repeating the previous step until the shooting of all the flight paths is completed.

Further, the flying speed of the unmanned aerial vehicle is fixed in the flying process of the unmanned aerial vehicle, and the shooting time interval is set according to the building size and the flying speed required to be shot.

Further, in the flight process of the unmanned aerial vehicle, the camera lens always faces the front of the unmanned aerial vehicle and vertically inclines downwards by 45 degrees.

Further, in step S3, for all photos taken in the flight process, photo scan is used to perform image preprocessing, feature point monitoring and matching, and dense point cloud generation on the photos; establishing a right-hand coordinate system by taking the center of the bottom of the building as an origin and the inclination direction of the building as the positive direction of an X axis to finish three-dimensional reconstruction; and reading and processing point cloud data through Python, and solving an interlayer displacement angle.

Further, the three-dimensional reconstruction method comprises the following steps:

and (5) performing three-dimensional reconstruction of the seismic damage structure on the screened photos by using Photoscan, and deriving point cloud data with three-dimensional coordinate axes.

Further, when it is detected that the inter-floor displacement angle of the building structure is greater than a prescribed value, the detection position is recorded and an alarm is issued.

Compared with the prior art, the invention has the beneficial effects that:

the invention can break through the limitation of the terrain and realize omnibearing and multi-angle detection; meanwhile, the probability of artificial misjudgment can be reduced by image identification; moreover, adopt unmanned aerial vehicle to shoot convenient, swift, detection time that can significantly reduce.

Drawings

Fig. 1 is a schematic circuit diagram of interlayer residual deformation detection proposed in the embodiment of the present invention;

FIG. 2 is an elevation view of a three-dimensional reconstruction based on a captured photograph in an embodiment of the present invention;

fig. 3 is a schematic circuit diagram of a detected building according to an embodiment of the present invention.

Fig. 4 is a flow chart of steps of the earthquake damage structure interlayer residual deformation detection method based on unmanned aerial vehicle camera shooting.

The specific implementation mode is as follows:

in order to make the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention are described in detail below with reference to the accompanying drawings.

Example (b):

in this embodiment, two-storey self-built concrete frame structure building is selected, the total height is 7.3 meters (4 meters for the first storey), the width of the frame is 1.8 meters, and the length of the frame is 2 meters.

A method for detecting residual deformation between earthquake damage structural layers based on unmanned aerial vehicle shooting, as shown in figure 4, comprises the following steps:

s1, enabling the unmanned aerial vehicle to fly according to the set flight route and shoot the building comprehensively;

as shown in fig. 1, the flight path includes 4 transverse arc flight paths, and 3 times of route switching operation is required; the starting position and the end position of each transverse arc-shaped flight path are the same; the arc-shaped flight routes are distributed at equal intervals from bottom to top;

fixing the flying speed and the shooting time interval at the initial position of each transverse arc flying route until the transverse arc flying route is finished; and controlling the unmanned aerial vehicle to move to the initial position of the transverse arc flight path of the previous layer at the end position of each transverse arc flight path, and repeating the previous step until the shooting of all the flight paths is completed.

In this embodiment, the technical requirements for shooting are as follows:

the shooting resolution was 2 mm. In the flight process of the unmanned aerial vehicle, the camera lens always faces the dead ahead of the unmanned aerial vehicle and is vertically inclined by 45 degrees, and is always perpendicular to the flight track of the unmanned aerial vehicle.

The maximum relative range is set to be H, the focal length is set to be f, the pixel size is set to be P, the shooting resolution is set to be G, the CMOS size width is set to be W, the CMOS size height is set to be V, the number of pixels corresponding to W is set to be the number of pixels corresponding to V.

In this embodiment, an intelligent aerial photography robot Phantom 4Pro in due, 8.8mm focus lens, with a CMOS size of 12.8 × 9.6mm and a picture size of 5163 × 3873 (pixels), is adopted, and a flight route is specifically designed according to the calculated parameters, as shown in fig. 3, the unmanned aerial vehicle is controlled to continuously shoot the unmanned aerial vehicle to take off right in front of the building according to the sequence of the flight route a → B → C → D → a → E → F → G → H → I → J → K → L → I, to ascend to 3 meters, to hover right in front of the point a, and to keep the distance from the tower body at about 2m, and the camera lens always faces right in front of the unmanned aerial vehicle and vertically inclines downward by 45 degrees. First 2 pictures were taken at the suspension point. Setting a flying distance of 18.8m around a circle, setting the flying speed to be 0.8m/s and setting the shooting interval to be 1.5s, creating a flying task, automatically completing the shooting task from A to B by the unmanned aerial vehicle, shooting 4 pictures, hovering at a point B, and then shooting 2 pictures. And manually operating the unmanned aerial vehicle to fly to the point C from the point B and hover, and repeating the previous step until 2 times of longitudinal flight lines are completed. After the above operation is completed, the parameters are reset, and the operation is reversed once, i.e., the flight path is I → L → K → J → I → E → H → G → F → E → a → D → C → B → a. And the method ensures rich image sources and is beneficial to three-dimensional reconstruction. During the whole flight, the transverse flying speed was kept at 0.8m/s, and pictures were taken at intervals of 1.5 s. The vertical direction is totally 6 times, and the distance of each time is 3m and totally 18 m. The full flight time is within 320 s. The number of pictures taken was 144 in total.

S2, transmitting the shot picture to a ground control platform;

and S3, after receiving the picture, the ground control platform carries out three-dimensional reconstruction on the shot building by using an image recognition technology and a three-dimensional reconstruction technology, calculates the interlayer displacement angle of the building and finishes the interlayer residual deformation detection of the seismic damage structure.

For all pictures shot in the flight process, carrying out image preprocessing, feature point monitoring and matching and dense point cloud generation on the pictures according to the shooting sequence; as shown in fig. 2, a right-hand coordinate system is established with the center of the bottom of the building as the origin and the inclination direction of the building as the positive direction of the X axis, the photos subjected to screening processing are subjected to three-dimensional reconstruction of the seismic damage structure, and point cloud data with three-dimensional coordinate axes are derived; and reading and processing point cloud data through Python, solving an interlayer displacement angle, recording a detection position and giving an alarm when detecting that the interlayer displacement angle of the structure in the building is larger than a specified value.

In this embodiment, each photo obtained should include a complete image of the plane and the side elevation.