An Efficient Salt-and-Pepper Noise Removal on Local Edge-Preserving Function

This paper presents a novel procedure for the salt-and-pepper noise remove on the local edge-preserving function. In the first phase, adaptive median filter is used to detect noisy pixels. In the second phase, local edge-preserving function is built. In the third phase, a minimization problem is solved by the Poly-Ribiere-Polak (PRP) method. Through the improvement on minimum allowed window size and maximum allowed window size of the adaptive median filter in the first phase, our method is much less running time in the phase. when the noise ratio is high, Comparing the results of removing noise, our method is better because of the local edge-preserving regularization, at the same time it is the PRP method by which the minimization problem is solved that keep our method robust and courage.