Present denoising algorithms of salt and pepper noise have many problems, such as high time complexity, neglect of the image edge and drawbacks of the details management. In order to get rid of these problems, this paper presents an algorithm of denoising salt and pepper noise based on edge classification. The algorithm first applies mesh division to denoising algorithms of salt and pepper noise. The edge mesh and non- edge mesh are processed by different methods. Experimental results showed the proposed algorithm could obtain higher Peak Signal-to- Noise-Ratio value and protect the edges of images. The research in this paper would have great significance to improve the quality of image.