International Journal of Computing Science and Information Technology (IJCSIT)
Impulse noise detection is a critical issue when removing impulse noise and impulse/gaussian mixed noise. The framework combines Robust Outlyingness Ratio (ROR) detection mechanism and Fuzzy C Means (FCM) clustering algorithm and Non-Local Means (NLM) filter. ROR for measuring how impulse like each pixel is and then all pixels are divided into four clusters according to the ROR values. The detection mechanism consists of coarse and fine stage. The output of coarse and fine stage of ROR is passed to FCM in which the authors use an iterative approach where each data point belongs to one or more centroids.