IMPULSE NOISE REMOVAL TECHNIQUE BASED ON FUZZY LOGIC FOR CONTRAST ENHANCEMENT
Mithilesh Atulkar, A. S. Zadgaonkar, S. Sasi Kumar · Transactions on Machine Learning and Artificial Intelligence · 2014
This paper presents an algorithm to remove random impulse noise in a digital grayscale image based on fuzzy logic approach. Numbers of filters are used to remove noise using fuzzy rules. Filter window used the local statistic to determine local weighted mean. Pixel that are detected noisy are filtered, other pixels are remain unchanged. Experimental results show that this method has better performance than other state of the art filters. The effectiveness of this method is 44.95 in terms of peak-signal-to-noise-ratio.