Abstract: A new change detection algorithm of polarimetric SAR image based on joint weighted polarization difference degree is proposed in this paper. First, the scattering difference degree and the power difference degree are computed between the bi-temporal POLSAR images. Secondly, weighted coefficients are chosen according to the relative size between scattering difference degree and the power difference degree. Then we created the difference map based on the joint weighted polarization difference degree. At last, the technology of threshold segmentation is used to extract the changed areas from the difference map. Two sets of fully polarimetric SAR datas collected by U.S. UAVSAR system are adopted to test and verify the new algorithm. Experimental results show that the novel method can find the changes of the surface features effectively with low false alarm rates and sharp outlines. Besides, this method can be used in any case without determining the statistical distribution of the image.