Rapid orthoimage generation method based on an unmanned aerial vehicle

Abstract

The invention discloses a fast ortho-image generation method based on an unmanned aerial vehicle. The method comprises the steps that images are clustered by integrating the similarity of scene content and space; Rapidly generating a sparse three-dimensional reconstruction result of each group on the basis of scene clustering, wherein the sparse three-dimensional reconstruction of each group is executed in parallel; Carrying out geographic registration under the condition of no ground control point, and fusing registration results of each group; Calculating an orthoimage corresponding to eachimage by utilizing similarity transformation, and finally superposing the orthoimages into an integral orthoimage. According to the method, matching of all the images can be completed with the lineartime complexity, and the matching relation between the images can be rapidly obtained; The number of optimization parameters of a single scene can be greatly reduced, so that the reconstruction efficiency is greatly improved; Accurate geographic registration can be realized only by utilizing the POS information obtained when the unmanned aerial vehicle flies and the camera position calculated by sparse three-dimensional reconstruction, and no additional control point is needed.

Description

Technical field [0001] The invention belongs to the technical field of aerial image processing, and more specifically, relates to a method for generating fast orthophotos based on drones. Background technique [0002] Compared with traditional aerial photogrammetry technology, UAV low-altitude remote sensing technology has the advantages of low cost, high resolution, and flexibility. However, the images obtained by UAVs are different from those obtained by traditional aerial photogrammetry. There are problems such as small image size and large quantity. These problems bring great difficulties to image matching and rapid image reconstruction. [0003] In the three-dimensional reconstruction technology based on image feature matching, image matching usually adopts a pairwise matching method, which brings great time consumption. Although it is possible to perform matching based on the sequence relationship of the images, due to the various route types of aerial photography, including...

AI Technical Summary

Problems solved by technology

Usually, many control points are required, so the workload of the field survey is very large. In addition, in areas with complex terrain or difficult conditions, it is difficult to manually set the control points

Images