Noise removal from image data using recursive neurofuzzy filters

Neurofuzzy approaches are very promising for nonlinear filtering of noisy images. An original network topology is presented in this work to cope with different noise distributions and mixed noise as well. The multiple-output structure is based on recursive processing. It is able to adapt the filtering action to different kinds of corrupting noise. Fuzzy reasoning embedded into the network structure aims at reducing errors when fine details are processed. Genetic learning yields the appropriate set of network parameters from a collection of training data. Experimental results show that the proposed neurofuzzy technique is very effective and performs significantly better than well-known conventional methods in the literature