Parallel Peer Group Filter for Impulse Denoising in Digital Images on GPU

keywords: Impulse denoising, color images, CUDA, parallel architecture, fuzzy metrics
A new two-steps impulsive noise parallel Peer Group filter for color images using Compute Unified Device Architecture (CUDA) on a graphic card is proposed. It consists of two steps: impulsive noise detection, which uses a Fuzzy Metric as a distance criterion and a filtering step. For the needed ordering algorithm we are using the Marginal Median Filter with forgetful selection sort. Comparisons with other color filters for Graphics Processing Unit (GPU) architectures are presented, demonstrating that our proposal presents better performance in color preservation and noise suppression.
reference: Vol. 38, 2019, No. 6, pp. 1320–1340