New image filtering algorithm based on impulse noise detection

For the limitation of the traditional impulse noise detection, an efficient method based on the image block iterative algorithm is presented in this paper. By the analysis of the image partial information, the image local optimal threshold can be got according to the similarity of local features. Then the noise can be extracted from the background which is made up of original image. So the purpose of eliminating noise can be achieved. The simulation results show that this algorithm is effectively to remove impulse noise from the image as well as reserve more details of the original image. And compared with the traditional image de-noising method, the proposed method produces a higher signal noise ratio.