Improved median filtering algorithm based on two-stage noise detection

To improve the performance of high salt and pepper noise reduction, a new improved filtering algorithm is proposed. First, a noise detection scheme which has two stages is adopted to identify the signal pixels and noise pixels. Second, the noise pixels are re- moved using the improved median filtering, while signal pixels hold their gray value and are left unprocessed. Simulation results show that the proposed algorithm can suppress noise very effectively and preserve image details very well, meanwhile, the restored images that are obtained by the proposed algorithm have better objective quality and subjective vision effect than that by the conventional median filter and other improved median filters.