Method and device for detecting change of SAR images

Abstract

The invention is applicable to the field of remote sensing and provides a method and device for detecting the change of SAR images. The method includes the steps of preprocessing SAR images of two periods; generating differential images according to the preprocessed SAR images of two periods; generating a fuzzy membership function Uu of an unchanged class [omega]u and a fuzzy membership function Uc of a changing class [omega]c through the differential images by using a fuzzy local information C-means clustering algorithm with adaptive distance; dividing the unchanged class [omega]u into the interior and fuzzy boundaries according to the fuzzy membership function Uu of the unchanged class [omega]u and dividing the changed classes [omega]c into the interior and fuzzy boundaries according tothe fuzzy membership function Uc of the changing class [omega]c; and de-fuzzifying pixels in the interiors and fuzzy boundaries of the unchanged class [omega]u and the changing class [omega]c respectively. According to the invention, the accuracy of the fuzzy membership function in the SAR image change detection and the classification accuracy of the fuzzy boundary pixels are improved and thus a relatively good SAR image change detection result can be obtained.

Description

technical field [0001] The invention belongs to the field of remote sensing, and in particular relates to a SAR image change detection method and device. Background technique [0002] Change detection in remote sensing images is a key technology in Earth observation applications such as agricultural surveys, disaster monitoring, urban studies, and resource monitoring. Since Synthetic Aperture Radar (SAR) has imaging advantages such as all-weather, all-weather and penetration performance on some ground objects, SAR image change detection technology has received more and more attention in recent years. [0003] Among them, the fuzzy clustering algorithm is one of the most widely used SAR image change detection techniques. However, fuzzy clustering SAR image change detection technology has the following defects: [0004] 1. In the process of generating the fuzzy membership function, the shape information of the unchanged class and the changed class is not considered, and the ...

AI Technical Summary

Problems solved by technology

[0004] 1. In the process of generating the fuzzy membership function, the shape information of the unchanged class and the changed class is not considered, and the statistical characteristics of the adaptive difference image cannot be adaptive, which affects the accuracy of the fuzzy membership function

[0005] 2. When using the maximum membership degree principle to defuzzify the fuzzy membership degree function, the difference of the fuzzy membership degree distribution of different pixels is not considered

For example, if the fuzzy membership degrees of pixel A and pixel B belonging to the changed class are 0.98 and 0.52 respectively, then the fuzzy membership degrees belonging to the unchanged class are 0.02 and 0.48 respectively, then according to the principle of maximum membership degree, both pixels A and B are marked is a change class, but for pixel B, its fuzzy membership degrees belonging to the change class and the unchanged class are very close, and the classification result has high uncertainty, which affects the accuracy of SAR image change detection

Images