遥感影像变化检测方法探讨

Review on change detection methods of remote sensing images

利用同一区域不同时相的遥感影像进行变化检测的应用一直是遥感应用研究的热点之一。本文对遥感变化检测技术的研究现状和新进展进行了一定的归纳和总结,对以往分类方法进行了介绍,分别对基于像素、基于对象以及结合特定理论的变化检测方法进行了重点讨论,并对每种方法涉及的理论和最新研究进展进行了评述。

The application of change detection is one of the hot spots in remote sensing application research. Change detection is a process that analyzes a pair of remote sensing images acquired on the same geographical area at different times in order to identify changes. In this paper, the research actuality and new process in change detection were introduced summarily. Firstly, the traditional classification method of change detection was introduced. Then, the change detection approaches separately based on pixel, object and combining with specific theory （ including Neural Networks, Support Vector Machine, Morphological Theory, Fuzzy Entropy Principle, Wavelet Transform, Swarm Algorithm, Genetic Algorithm ） were presented emphatically, and the involved theory and the latest re- search were briefly reviewed. Finally, remote sensing change detection methods were discussed.