A synthetical pose estimation of SAR imagery using Hough transform and 2-D continuous wavelet transform

SAR target recognition close relates to target pose. When the pose of target is confirmed, the computational complexity can be deduced, which can increase the recognition probability greatly. Hough transform is an efficient approach to recognize the target pose. But, in general, Hough transform can obtain many poses for a single SAR target. The right pose can't be obtained only using Hough transform. In order to resolve this problem, in this paper, the angular energy density of 2-D continuous wavelet transform combined with Hough Transform is proposed. Many experimental results of MSTAR public target database show that this method is feasible and it can obtain the pose correctly.