Image denoising searching similar blocks along edge directions

The block-matching and 3D filtering (BM3D) method uses the fixed directions to search for similar blocks over the entire image; the performance of this method may be limited when handling edges. In this study, we propose an approach that searches for candidate matching blocks along the edge directions that are well adapted to image details. All similar blocks form a 3D group. Denoising is performed by shrinking the coefficients of the 3D transform applied on these groups. The experimental results show that the proposed method outperforms some of the current denoising methods in terms of numerical results and visual quality; furthermore, the proposed method introduces only a few artifacts while preserving edges and textures.