Multichannel information processing from a diverse range of channel information is highly time-and space-complex owing to the variety and enormity of underlying data. Most of the classical approaches rely on filtering and statistical techniques. Methods in this direction involve Markov random models, vector directional filters and statistical mixture models like Gaussian and Dirichlet mixtures. The non-classical approaches comprising the neuro-fuzzy-genetic paradigm or its variants are bestowed

more »

... with features for real time applications. This article presents a brief survey of the aforestated trends in color image enhancement and segmentation.