Accurate segmentation of noisy images using anisotropic diffusion and linking of boundary edge
J. Maeda, Tomoya Iizawa, Tohru Ishizaka, Yudai Suzuki · 2002
We present a segmentation of noisy images that uses an anisotropic diffusion algorithm as a precise edge-preserving smoothing technique. We propose a modified version of the anisotropic diffusion algorithm by using the boundary edge. We incorporate a linking algorithm for the boundary edge by a directional potential function to improve the ability of edge-preserving smoothing. As a result, the unnecessary details of images are effectively smoothed before performing a region growing algorithm as a segmentation procedure. Therefore, the proposed method is suitable for accurate segmentation of noisy images. Several simulated examples are presented using the proposed method that demonstrate the effectiveness of the proposed technique.