Image Matching Method Based on Robust Feature Matching of Advanced Neighborhood Topological Consistency

Abstract

The invention relates to an image matching method based on robust feature matching of advanced neighborhood topology consistency. According to the Bayesian principle, the method establishes the guided matching strategy model between the matching images A and B, establishes the soft assignment expression between the length ratio of the vectors in the neighborhood where the images A and B are located and the angle between them, and establishes the soft assignment expression between the neighborhood topology according to the neighborhood topology. The weighted consistency is used to judge the smoothness and construct a mathematical model seeking the minimum cost according to the length ratio and included angle of the selected vectors in the two images and the method of multi-scale K neighborhood. The image matching method of the present invention has good performance in the three indicators of precision, recall and F value, so the image matching method of the present invention has a good application prospect, and can be applied to three-dimensional reconstruction and to quickly realize the splicing of remote sensing images of UAVs and other fields.

Description

technical field [0001] The invention relates to the field of image matching, in particular, the invention relates to an image matching method based on robust feature matching based on advanced neighborhood topology consistency. Background technique [0002] For a long time in the past, feature matching has developed into the most important form of image matching. Since it aims to identify and pair salient and stable feature structures with the same or similar properties from two images, it is a fundamental and critical problem in the fields of computer vision, pattern recognition and image processing. However, feature matching is essentially a complex combinatorial optimization problem with extremely high computational complexity. At the same time, the influence of noise, outliers and complex scene transformations cause existing feature matching methods to have many deficiencies, which seriously restricts its practical application capability. Therefore, researching an accu...

AI Technical Summary

Problems solved by technology

However, the main disadvantage of LPM is that the consistency of its existing neighborhood topology cannot distinguish inliers from outliers well, and it is easy to excessively penalize outliers

In addition, when the proportion of outliers in the initial match is high, the construction of its neighborhood structure will no longer be reliable

Images