We present a novel nonlocal mean (NLM) algorithm using an anisotropic structure tensor to achieve higher accuracy of imaging\r\ndenoising and better preservation of fine image details. Instead of using the intensity to identify the pixel, the proposed algorithm\r\nuses the structure tensor to characterize the boundary information around the pixel more comprehensively.Meanwhile, similarity\r\nof the structure tensor is computed in a Riemannian space for more rigorous comparison, and the similarity weight of the pixel (or\r\npatch) is determined by the intensity and structure tensor simultaneously. The proposed algorithm is compared with the original\r\nNLM algorithm and a modified NLM algorithm that is based on the principle component analysis. Quantitative and qualitative\r\ncomparisons of the three NLM algorithms are presented as well.
Loading....