This paper proposes a novel adaptive neuro-fuzzy inference system (ANFIS) based impulse detection method for the restoration of images corrupted by impulse noise (IN). After the corrupted pixels detected by proposed detector, the Median filtering is performed for only these pixels. The performance of the proposed neuro-fuzzy detector based median filter (NFDMF) is evaluated on different test images and compared with 14 different comparison filters from the literature. Experimental results show that the proposed filter shows better performance than the comparison filters in the cases of being effective in noise suppression and detail preservation, especially when the noise ratio is very high. (C) 2010 Elsevier GmbH. All rights reserved.