A 2-D adaptive least lp-norm filter for impulsive noise cancellation in still images

In this paper, we propose a new class of filters based on fractional lower order statistics (FLOS) for still image restoration in the presence of a-stable noise signals. For this purpose, we present a family of 2-D finite-impulse response (FIR) adaptive filters optimized by the least mean l_p-norm algorithm. Experiments performed on natural images indicate that the proposed algorithm provides superior performance in impulsive noise environments compared to existing approaches.