A fast three-dimensional image reconstruction method and device

Abstract

The invention discloses a method and device for fast three-dimensional reconstruction of stereoscopic image are disclosed. The method comprises the steps of acquiring a frame-shaped sequence image of a stereoscopic observation area; conducting dense image point homonymy matching of the frame-shaped sequence image; determining a spatial reference plane and internal and external reference parameters of a stereoscopic image in a homonymous point set; obtaining the out-of-plane distance from each point in the homonymous point set to the spatial reference plane according to the homonymous point set, the spatial reference plane and the internal and external reference parameters of the stereoscopic image; generating three-dimensional point clouds according to the out-of-plane distance from each point in the homonymous point set to the spatial reference plane; and reconstructing the three-dimensional topography of the stereoscopic observation area according to the three-dimensional point clouds. The invention can significantly reduce the calculation amount and the calculation difficulty without affecting the calculation result, and improves the calculation efficiency, so that it is possible to reconstruct the three-dimensional point clouds rapidly.

Description

technical field [0001] The present invention relates to image three-dimensional reconstruction technology, in particular, to a fast three-dimensional image reconstruction method and device. Background technique [0002] In the independent research and development of digital city technology, 3D image reconstruction is a core task, which mainly refers to the 3D reconstruction of the target scene based on the frame sequence images obtained by ground mobile collection, drone aerial photography, etc.; The amount of urban work data is huge and the calculation is complex. How to quickly and efficiently perform image-based 3D reconstruction of the target scene has become a key issue. Research on various 3D reconstructions in computer vision generally focuses on the matching of image homonym points. As for how to reconstruct 3D point clouds from the matched dense homonym points, it is not the focus of research. However, as the amount of image data is geometrically Increase, the amou...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the problem in the prior art that reconstructing a 3D point cloud with densely matched points with the same name is too computationally intensive when the amount of image data is large, the purpose of the present invention is to propose a fast 3D reconstruction method and device for a stereoscopic image, which can change the matching The calculation method of reconstructing the 3D point cloud with the dense homonym points after that can greatly reduce the amount of calculation actually generated when the amount of image data is large, and greatly reduce the calculation processing time, so that the method of reconstructing the 3D point cloud has practical application value

Images