SIR-C is the first space-borne imaging radar system with multi-wavelength and quad-polarization developed by the joint effort of the USA, Italy and Germany. Polarization SAR measures the scattering matrix of each pixel on ground and synthesizes the image at given orientation and ellipticity angle, including linear and elliptical polarization. It has many advantages over single or multi-polarization SAR in such aspects as detecting objects, identifying targets and extracting texture. Nevertheless, the relatively high correlation of the synthesized polarized images and the complexness of scattering of objects often lead to wrong interpretation of the images and poor accuracy of classification. Based on SIR-C data of Hetian prefecture in Xinjiang, the authors used the target decomposition theory to decompose the data into three non-relevant scattering components. The result shows that the decomposed three scattering components reflect the correct scattering feature. The authors then combined them with polarimetric synthesized SAR power image to classify the experimental area by using MLC or neural net. The classification result shows that the method can effectively extract the information of land cover, achieve relatively good classification accuracy of ground objects and improve the capability of SAR for monitoring the land use and cover.