Decision Tree Based Algorithm for Removal of Different Probability Salt and Pepper Noise in Images
International Journal of Recent Technology and Engineering (IJRTE) · 2019
A Decision tree based algorithm for the removal of equal and unequal probability salt and pepper noise in images is proposed. The algorithm aims to address one solution for different salt and pepper noise models. The proposed algorithm operates on an image using fixed 3*3 window. The decision tree based algorithm classifies pixel into noisy or not based on the decision and replaces it with mean of neighbours or unsymmetrical trimmed median or unsymmetrical trimmed midpoint. The algorithm exhibit excellent noise elimination capability at high noise densities in terms of quantitative and qualitative perspective. The algorithm was found to exhibit good noise removal characteristics for three noise models