泛概念层次构建方法及其在遥感图像分类中的应用

Pan-concept-level Generation Method and Its Application in Remote Sensing Image Classification

针对传统泛概念树在进行正态云综合时出现雾化现象以及使用极大判定法进行遥感图像分类时缺乏类别信息的问题,提出了一种泛概念层次构建的改进算法,并基于改进的泛概念层次构建算法进行遥感图像分类。对TM遥感影像进行分类实验,并与常用的最大似然、最小距离、马式距离等方法进行遥感图像分类的对比试验,验证了本方法的有效性。

The paper discusses a new pan-concept-level generation method based onthe cloud model. A new method of remote sensing image classification based on a improved pan-con- cept-level generation method is proposed. Classification experiments are used to compare the proposed method with traditional methods. Comparative experiments validate the proposed method.