Abstract: A novel speckle reduction algorithm based on joint sparse representation of multipolarization synthetic aperture radar (SAR) image is presented. First, the data of SAR image are used to train the dictionary. Then joint sparse model is constructed. Finally, joint sparse coefficient is solved with orthogonal matching pursuit algorithms, to reconstruct the clean images of each channel . Experimental results with the data of hayward area from AIRSAR system show that the proposed method is effective both on speckle reduction and strong scattering point target signature preservation compared with the synthetic power image results of each channel image sparse denoising separately.