高分辨率遥感影像多特征协同地物分类方法

High Resolution Remotely Sensed Image Classification Method of Multi-Feature Collaboration

高空间分辨率遥感影像在提高对地物细节信息表达的同时,因地物类内光谱方差增大、类间光谱方差降低而造成了影像上地物识别与分类难度的增加。针对高分辨率遥感影像的地物分类问题,提出并实现了一种将光谱、纹理、形状等多特征综合协同的分类方法,并通过实验验证了该方法的有效性。

High resolution remotely sensed images can provide a large amount of detailed information. However, due to the increasing variance within the same class and the decreasing variance even between different classes in spectral domain, the difficulty of land cover＇s recognition and classification increases a lot. In order to overcome the problem, a classification method of combing spectrum, texture and shape features into collaborative features was proposed. The experiment results showed that the proposed method was effective.