On the reduction of impulsive noise in multichannel image processing

A new approach to the problem of impulsive-noise reduction for color images is introduced. The major advantage of the technique is that it filters out the noise component while adapting itself to the local image structures. I n this way the algorithm is able to eliminate strong impulsive noise while preserving edges and fine image details. As the algorithm is a fuzzy modifi cation of the commonly used vector median operator, it is very fast and eas y to implement. Our results show that the proposed method outperforms all s tandard algorithms for the reduction of impulsive noise in color images. (C ) 2001 Society of Photo-Optical Instrumentation Engineers.