AEU-INTERNATIONAL JOURNAL OF ELECTRONICS AND COMMUNICATIONS, cilt.58, sa.1, ss.72-75, 2004 (SCI-Expanded)
A new neuro-fuzzy edge detector for digital images corrupted by impulse noise is presented. The structure of the detector is very simple and comprises four identical neuro-fuzzy subdetectors and a postprocessor. The internal parameters of the detector are determined by training. The detector efficiently extracts edges in digital images corrupted by impulse noise without requiring the filtering of the noise.