SAR image registration method based on line and area

Abstract

The invention discloses an SAR (specific absorption rate) image registration method based on straight lines and area, and mainly solves the problems of poor registering effect and high calculating complexity in the prior art. The SAR image registration method including the implementing steps of (1) inputting two SAR images and detecting straight lines in the two images through Hough conversion; (2), grouping the detected straight lines according to slope to acquire principal directions of the two SAR images, and calculating rotary parameter; (3), processing standard images and rotated images to be registered by the aid of an optical threshold segmentation method and through morphology corrosion and expansion to acquire a closed area; (4), utilizing an area centroid to calculate translation parameters; and (5), coinciding the standard images and the translationally rotated images to be registered to finish registration. The SAR image registration method based on the straight lines and area has the advantages that the SAR image registration method has small noise effect on registration of SAR images, is stable in SAR image segmentation results in a small water area and good in registration effect, high in speed and low in calculating complexity, and can be used for mode recognition, automatic navigation, computer vision and remote sensing image processing.

Description

technical field [0001] The invention belongs to the technical field of image processing, relates to synthetic aperture radar SAR image segmentation, and can be used for SAR image registration. Background technique [0002] Synthetic Aperture Radar (SAR) is a high-resolution active microwave remote sensing imaging radar, which has the advantages of all-weather, all-time, high resolution, and side-view imaging. It can be used in military affairs, agriculture, navigation, geographic surveillance, etc. field. Image matching can be divided into three categories: matching methods based on gray correlation, matching methods based on features and matching methods based on mutual information. [0003] The method based on gray-scale correlation is a way to search and match images with a certain size of Dido array in order of one or several similarity measures. Once this kind of matching enters the area of ​​poor information or large scale difference or distortion of the image, the m...

AI Technical Summary

Problems solved by technology

[0005] The template matching algorithm based on mutual information has the advantages of high matching accuracy, anti-occlusion, anti-random noise, etc., and has been widely used in the field of image registration, but the calculation amount of standard mutual information is very large, and it is difficult to achieve real-time matching; and, mutual information The measurement loses a lot of image space information, which will lead to problems such as the maximum mutual information is not unique or the maximum position is inconsistent with the image registration position.

Images