Maximum histogram different-source image matching method based on point-line features

Abstract

The invention discloses a maximum histogram different-source image matching method based on point-line features. A position-based boosting detector is adopted to extract a large number of feature points distributed on a thick edge, and a local maximum edge direction and a gradient direction of the feature points are selected from combined features to construct histogram features, so that the problem of time consumption of dense matching is expected to be solved; feature points of initial matching are extracted based on bilateral matching, the common mismatching problem in matching is solved byadopting the thought based on differential variance, and the matching accuracy is improved. What is ensured is that the matching system works in an uncertain environment, objective evaluation is carried out on the proposed different-source image matching algorithm, and accurate and densely matched matching point pairs can be stably output.

Description

technical field [0001] The invention belongs to the technical field of image processing, in particular to a maximum histogram heterogeneous image matching method based on point-line features. Background technique [0002] In recent years, heterogeneous image analysis has been widely used in many fields such as tracking, video surveillance, guidance and remote sensing monitoring. These applications are often images acquired by sensors using different imaging mechanisms. The so-called heterogeneous image matching is the process of locating the same features and regions from different sensor images. The information provided by heterogeneous image pairs can complement each other, which is beneficial to solve various detection problems in the absence of effective information. In these vision tasks that require image analysis, image matching is one of the most important processing steps. However, due to the different principles of image acquisition by different sensors, the acq...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a maximum histogram heterogeneous image matching method based on point-line features. By matching different imaging features under different sensors, the useful information of each band is fused to achieve all-weather monitoring and accurate positioning of remote sensing images, and fully analyze The impact of image features on heterogeneous image matching systems, establishing the characteristic elements that affect the performance of the matching system, designing a fusion of local feature descriptors, and solving the problem of single feature matching methods in the matching of heterogeneous images with large differences in regional features. unstable problem

Images