Efficient Non-linear Filter for Impulse Noise Removal in Document Images

A novel method is proposed in this paper to restore document images. The proposed method is based on finding the degree of similarity between the tested pixel and its neighbors in different door's sizes. If the tested pixel in every door size has enough similarity with at least two pixels, then the tested pixel is deemed original pixel. The number of two pixels is chosen, to make sure that the tested pixel is a part of an original text or a part in a series of similar pixels. Simulation results indicate that the new method delivers superior performance rapidly and efficiently either in terms of the noise removal or details preservation.