Tilt photography three-dimensional modeling

Before it is too late, let's introduce the technical system of oblique photography three-dimensional modeling.

1, overview

With the development of geographical technology, geographical application is undergoing a transformation from traditional two-dimensional to three-dimensional. Three-dimensional modeling of cities has also developed from traditional manual modeling to low-cost large-scale modeling based on tilt photography and institutional radar.

The complete oblique photography 3D modeling includes three parts: data acquisition, modeling and application. Data collection involves UAV aerial photography, camera, route planning, data quality control and so on. Modeling includes three-dimensional operation, reconstruction and model modification; The application part is mainly the combination of three-dimensional model and specific business, and the common functions include earthwork calculation, building height limit analysis, intelligent city with rooms to check people, and patient rounds with people. The overall technical system is very extensive. With the help of project practice, I was lucky enough to experience all parts of modeling. Due to time and technical reasons, I have a little understanding of each part, so I wrote it down as a summary.

2. What is oblique photography?

Building surface images are obtained from multiple sides, and the internal and external orientation elements of the camera position are known. According to the principle of rear intersection, the spatial position of the photographed object is determined, and the three-dimensional modeling is realized on this basis. Tilt photography generally obtains data from five directions, namely vertical orthogonal projection and four directions tilt (generally 45 degrees). The internal orientation elements of the camera include the focal length and the position of the main point of the image, while the external orientation elements include the spatial position and attitude information of the camera. The rear intersection is the basic method of triangulation.

Flight platform: it mainly includes fixed wing and rotor. The fixed wing has a long flight time and the rotor is relatively short. The DJI M600 pro used in my project has a set of 6 batteries, and the flight time in plateau area is about 20 minutes.

Camera: Common schemes include five lenses, double lens swing and single lens. As long as the theory satisfies the triangulation principle, the modeling can be completed by measuring the surface of the same object from at least two different positions. So for a single lens, it usually takes five flights to get the data of five routes (five angles of the building).

3. Data acquisition

4. Modeling

5. Application

6. Summary