The thesis presents a novel adaptive class of decision based non-linear rank-order filter operators equipped with a variety of distinct characteristics to deal salt & pepper impulse noise situations in digital images even those which consume 95% signal content. The filters more efficiently choose an appropriate impulse restorer by adaptively moving over an acceptable image space to retain sharp details with least blocky, patchy effects.
Published in:
Advanced Computing and Communications, 2006. ADCOM 2006. International Conference on
Date of Conference: 20-23 Dec. 2006