Inclined image relative orientation method based on essential matrix decomposition

Abstract

The invention relates to an inclined image relative orientation method based on essential matrix decomposition. The inclined image relative orientation method is characterized by comprising: calculating an essential matrix E; recovering a camera matrix P from the essential matrix; converting the camera matrix P into consecutive relative orientation element initial values; converting the consecutive relative orientation element initial values into independent orientation element initial values; and constructing an independent relative orientation error equation, solving the independent relativeorientation element correction value through a least square method, updating the independent relative orientation element, and carrying out least square iteration solving so as to achieve the inclined image relative orientation function. According to the present invention, with the relative orientation method comprising obtaining the consecutive relative orientation element initial values based on the essential matrix decomposition and performing using least square iteration optimizing, the problem that the traditional relative orientation method cannot be adapted to the inclined image relative orientation is solved, and the precision requirement of the photogrammetry on the relative orientation is met; and the method of the present invention is suitable for the inclined aerial photography images with large relative attitude and any relative positions.

Description

technical field [0001] The invention belongs to the technical field of photogrammetry and remote sensing, and relates to the technology of relative orientation of oblique photogrammetry images, in particular to a relative orientation method of oblique images based on essential matrix decomposition. Background technique [0002] At present, in the post-processing of oblique photogrammetry data, there is a problem of relative orientation of images with large relative poses and arbitrary relative positions. The relative orientation method of traditional photogrammetry, because of its small relative attitude and fixed relative position, uses empirical values ​​such as 0 as the initial value, and performs least squares iterative solution. This method is obviously not used to solve the relative orientation of oblique images question. In the field of computer vision, the relative orientation problem can be described as restoring the relative position and attitude between the two c...

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to overcome the deficiencies of the prior art and provide a method for relative orientation of oblique images based on essential matrix decomposition to solve the problem that traditional relative orientation methods cannot adapt to the relative orientation of oblique images

Images