Mixed Noise Suppression in Color Images by Signal-Dependent LMS L-Filters

Abstract

The paper is devoted to the signal-dependent (SD) design of adaptive LMS L-filters with marginal data ordering for color images. The same stem of SD processing of noised grayscale images was applied on noisy color images. Component-wise and multichannel modifications of SD LMS L-filter in R'G'B' (gamma corrected RGB signals) color space were developed. Both modifications for filtering two-dimensional static color images degraded by mixed noise consisting of additive Gaussian white noise and impulsive noise were used. Moreover, single-channel spatial impulse detectors as detectors of impulses and details were used, too. Considering experimental results, SD modifications of L-filters for noisy color images can be concluded to yield the best results.