论文部分内容阅读
为克服基于极化散射特性保持的迭代Wishart分类算法不适用于城区及对混合散射像素分类欠理想等不足,本文提出一种改进方法。其基本思想是先应用四分量分解算法将像素分成4种基本散射类型和混合散射类型,接着以平均合并度为指导对基本散射类型中的像素自适应聚类,最后对所有像素进行散射特性保持的迭代Wishart分类。试验结果表明,改进算法较原始算法有更好的适应性和分类效果。
In order to overcome the shortcomings that the iterative Wishart classification algorithm based on polarization scattering characteristics is not suitable for urban areas and the sub-ideal classification of hybrid scattering pixels, an improved method is proposed in this paper. The basic idea is to divide the pixels into four kinds of basic scattering types and mixed scattering types first by using the four-component decomposition algorithm, and then adaptively cluster the pixels in the basic scattering types with the average combining degree as the guideline, and finally, keep the scattering characteristics of all the pixels Iteration Wishart classification. The experimental results show that the improved algorithm has better adaptability and classification effect than the original one.