SAR image similarity measuring method based on point set contrast

Abstract

The invention discloses an SAR image similarity measuring method based on point set contrast. Firstly, two-dimensional scattering center extraction is carried out to reference SAR images to be matched and contrast SAR images, thus obtaining a reference scattering point set and a contrast scattering point set; the reference point set and the contrast point set are divided into a plurality of gathered point groups according to the distribution of the scattering points in the two-dimensional space; the coordinates of the two point sets are mapped to a same two-dimensional matching space; determining cross-image point group pairs between the reference point set and the contrast point set according to the distances among the point groups in the point sets; combining the distance similarity and gray similarity of every cross-image point group pair to obtain final similarity of the cross-image point group pair; and synthesizing the similarity of all cross-image point group pairs as the similarity among the SAR images. According to the method of the invention, the problem that the image similarity of the images obtained by the existing similarity judging method is inaccurate is solved; and the method of the invention is featured by simple realization steps, strong stability and little demanded calculation quantity and memory consumption.

Description

technical field [0001] The invention belongs to the fields of SAR image automatic interpretation and SAR image registration, and relates to a SAR image similarity measurement method based on point set comparison. Background technique [0002] SAR image similarity measurement aims to quantify the similarity between SAR images, so as to achieve the role of image screening and evaluation, and is mainly used in automatic interpretation of SAR images and image registration. It is an effective way to improve the accuracy and efficiency of image screening to study the similarity metrics with high reliability, high efficiency and compliance with visual perception. [0003] In the prior art, all pixel information of the SAR image is directly used to calculate the similarity measure, which has poor timeliness and low reliability. The strong scattering points of the SAR image represent the fine physical structure of the target. Taking them as the object of point set similarity measure...

AI Technical Summary

Problems solved by technology

However, although the algorithm is easy to implement, it is not robust

[0005] In 2000, Zhang Lihua’s doctoral thesis of Tsinghua University conducted a special study on point pattern matching, and gave a relatively common definition of matching, but did not build a theoretical framework for point set matching, and the similarity measurement problem was not discussed systematically. The specific application background of this method is relatively limited

Images