Impulse noise removal from color images using adaptive neuro-fuzzy impulse detector

Abstract

In this paper, we present a filtering scheme based on adaptive neuro-fuzzy inference system (ANFIS) for restoration of color images. The adaptive neuro-fuzzy impulse noise detector provides a reliable detection of noisy pixels in the noisy images. The filtering stage combines the best features of both component-based as well as vector-based approaches to restore the pixel value. The proposed scheme, thus, performs both brightness and color correction. Efficacy of the proposed method is verified through computer simulations which indicate that the present filtering scheme outperforms several other filters used in the study. © 2010 Springer-Verlag Berlin Heidelberg.