Parallel Fuzzy Filter for Impulse Noise Removal

A parallel algorithm for reducing impulse noise from color images is proposed. The algorithm is based on a fuzzy metric and is performed in two steps followed by noise filtering using the vector median filter. An implementation of the algorithm on multi-core interface using the Open MultiProcessing (OpenMP) is presented. A performance analysis with large images is conducted. Performance is evaluated in terms of execution time and in terms of PSNR. Results show that the proposed filter obtains good performance in terms of PSNR. After applying the multicore optimization strategies, the observed time shows that the proposed filter is able to remove impulse noise in real-time.