An improved parabolic-hyperbolic anisotropic diffusion algorithm for image denoising

Traditional noise reduction methods couldn´t well achieve the trade-off between removing noise and preserving feature information, inevitably blurring the meaningful details. Focusing on this problem, in this paper, we proposed an improved anisotropic diffusion parabolic-hyperbolic PDE for image denoising. Based on viewing the image as an elastic sheet and a modified anisotropic diffusion, the proposed algorithm took the gray-level variance information into consideration. We applied the proposed approach to several real images contaminated by white Gaussian noise with different standard deviation. The comparative experimental results show that the improved algorithm is capable of removing noise without sacrificing the edges and fine details of the image and therefore obtains superior denoising performance.