Abstract:

To the limitations of traditional land use interpretation technologies,the spectral features of ground object were analyzed deeply.And a classification method using spectral angle classification technique to classify the first-level land types and according the spectral angle image and second-class spectral characteristics to build classification rules and classify the second-class spectral characteristics was proposed.This method was used to interpret remote sensing image,and the classification accuracy of this method's classification results were compared with the maximum likelihood classification in the supervised classification.The results show that the method's classification results had better classification accuracy than the maximum likelihood classification.Its area precision and spatial accuracy was improved obviously.And this method can be used as the classification method of complex ground object.

Key words: remote sensing, land use, spectral angle classification