关键词: 图像去噪, 自蛇模型, 偏微分方程（PDE）, 非线性扩散

Abstract: According to the shortcoming of P-M nonlinear diffusion model and the self-snake model during the diffusion process, in order to utilize fully advantage of these two models, a novel image denoising method based on mixing self-snake model and P-M diffusion model is proposed, and a fidelity term is added into mixed model. Denoising and edge preserving process can be obtained to a good result. Finally, experiment results show that the proposed method not only remove noise efficiently, but also retain detail information well, such as edges.

Key words: image denoising, self-snake model, Partial Differential Equation（PDE）, nonlinear diffusion