CN112179320A - Method for creating 3D model by using Mini unmanned aerial vehicle in cooperation with common digital camera for shooting - Google Patents

Abstract

A method for creating a 3D model by using a Mini unmanned aerial vehicle in cooperation with common digital camera photography is characterized by comprising the following steps: the method comprises the steps of taking a picture by a Mini unmanned aerial vehicle carrying a digital camera, creating a 3D model by an external orientation element calculation method and a digital ground model (DSM) making method, flying in a specified area by a Mini type unmanned aerial vehicle carrying a digital camera, obtaining aerial photos and simultaneously recording the external orientation element of each photo, recovering the correlation between the aerial photos and the shot ground by obtaining six external orientation elements, rebuilding a ground stereo model by software after establishing the correlation between the photos and the shot ground, and extracting the geometric and physical information of a target by using the stereo model.

Description

Method for creating 3D model by using Mini unmanned aerial vehicle in cooperation with common digital camera for shooting

One, the technical field

The invention relates to a method for creating a 3D model, in particular to a method for creating a 3D model by using a Mini unmanned aerial vehicle and a common digital camera for shooting.

Second, technical background

With social progress, the traditional measurement mode faces factors of low efficiency, insufficient data volume, high production cost and the like, and causes the measurement industry to gradually become labor-intensive industry, thereby aggravating the development way that the measurement industry needs to seek new methods and new technologies. With the high-speed development of the unmanned aerial vehicle technology and the maturity of the photography technology and the GPS technology, the unmanned aerial vehicle has the advantages of low data acquisition cost, flexibility, high speed, high spatial resolution and the like, and is widely applied to the fields of national condition monitoring, geology, precision agriculture, environment monitoring, urban planning and the like. According to the difference of application fields, the unmanned aerial vehicle can be divided into a large-scale professional-grade unmanned aerial vehicle and a small-scale unmanned aerial vehicle. Although the measurement accuracy of large-scale professional unmanned aerial vehicle can reach centimeter level, the large-scale unmanned aerial vehicle has tedious work flow and great control difficulty, needs professional operation and must apply for flight airspace. Relatively speaking, the Mini type unmanned aerial vehicle has the advantages of easy operation, low cost, high efficiency, flexible flight and the like, and is gradually widely applied.

Third, the invention

The method comprises the steps of taking a Mini type unmanned aerial vehicle and a common digital camera as main tools, establishing a 3D model through cooperation of the unmanned aerial vehicle and the common digital camera and through an external orientation element calculation method and a Digital Surface Model (DSM) manufacturing method.

The method has the advantages that: mini type unmanned aerial vehicle is light, easily controls, and application scope is wide.

Fourthly, explanation of the attached drawings:

FIG. 1 is a schematic view of the positioning under the shielding condition of the present invention

The fifth embodiment is as follows:

the method for creating the 3D model by the Mini unmanned aerial vehicle in cooperation with the common digital camera comprises the following specific steps:

(1) oblique photogrammetry is carried out in a mode that a Mini type unmanned aerial vehicle carrying a digital camera spirals and rises around the tree;

(2) when oblique photogrammetry is carried out, the overlapping rate of more than 90% between adjacent images collected at the same height is ensured, the overlapping rate of more than 60% between images collected at different heights is ensured, the optimal angle of a camera during oblique photography is determined to be 45 degrees, more detailed and rich information can be obtained when the camera shoots a ground target tree at the angle, and the modeling effect is better;

(3) an attitude sensor carried by the unmanned aerial vehicle simultaneously records the attitude of the camera and records the external orientation elements of each photo;

(4) after the flight is finished, processing the acquired oblique image by utilizing professional software through the steps of air-to-three solution, homonymy point matching, dense point cloud generation and the like to generate point cloud data of a measuring area, and further completing the reconstruction of a three-dimensional point cloud model of the ground target object;

(5) the reconstruction result is output in a point cloud data format, mainly comprising LAS and XYZ; each point in the cloud data comprises three-dimensional coordinate information and color information; the coordinate information records the geometric position of a certain point on the target object, but does not reflect the topological relation between the point and the point, so that the recovery of the real shape of the target object needs to be realized by professional post-processing software; the color information records color values (RGB) of corresponding pixels of points in the point cloud on the oblique image; the accurate position of the target object can be determined by the three-dimensional coordinate information of the passing point, and the geometric parameters of the target object can be measured.

Claims (1)

1. A method for creating a 3D model by using a Mini unmanned aerial vehicle in cooperation with common digital camera photography is characterized by comprising the following steps: shooting by a digital camera carried by a Mini unmanned aerial vehicle, and creating a 3D model by using an external orientation element calculation method and a digital terrain model (DSM) making method, wherein the steps are as follows:

(1) oblique photogrammetry is carried out in a mode that a Mini type unmanned aerial vehicle carrying a digital camera spirals and rises around the tree;

(2) when oblique photogrammetry is carried out, the overlapping rate of more than 90% between adjacent images collected at the same height is ensured, the overlapping rate of more than 60% between images collected at different heights is ensured, the optimal angle of a camera during oblique photography is determined to be 45 degrees, more detailed and rich information can be obtained when the camera shoots a ground target tree at the angle, and the modeling effect is better;

(3) an attitude sensor carried by the unmanned aerial vehicle simultaneously records the attitude of the camera and records the external orientation elements of each photo;

(4) after the flight is finished, processing the acquired oblique image by utilizing professional software through the steps of air-to-three solution, homonymy point matching, dense point cloud generation and the like to generate point cloud data of a measuring area, and further completing the reconstruction of a three-dimensional point cloud model of the ground target object;

(5) the reconstruction result is output in a point cloud data format, mainly comprising LAS and XYZ; each point in the cloud data comprises three-dimensional coordinate information and color information; the coordinate information records the geometric position of a certain point on the target object, but does not reflect the topological relation between the point and the point, so that the recovery of the real shape of the target object needs to be realized by professional post-processing software; the color information records color values (RGB) of corresponding pixels of points in the point cloud on the oblique image; the accurate position of the target object can be determined by the three-dimensional coordinate information of the passing point, and the geometric parameters of the target object can be measured.

Images