Approach of SAR Image Classification Based on Wavelet Transform and FCM

摘要

The major task of synthetic aperture radar (SAR) is the classification of land cover. Wavelet transform (WT) has the advantage of multi-scale analysis, which is especially fit for classifying image texture. In this paper, a new method of SAR image classification is proposed The concrete step is as follows, first texture features are extracted with wavelet, SAR image is filtered according to the features of its distribution in the wavelet field, at last, this approach adopts the technology of texture classification based on fuzzy C-means clustering (FCM) algorithm. With the experiment result, the method is proved effective in SAR image classification.