An Efficient Weighted Median Filter for the Restoration of Images Corrupted by Random Valued Impulse Noise

Abstract

The paper proposes a new adaptive weighted median filter for the restoration of impule corrupted digital images by incorporating the limitations posed by Directional Weighted Median Filter (DWMF [4] and the algorithm proposed by Justin et. al [20]-[21]. The algorithm incorporates sixteen directions to identify the presence of edges and to preserve these detected edges while restoring images. The proposed filter uses an adaptive iterative median filter to replace the impulses. The experimental results shows that the proposed filter works better than other competitive filters and is capable of preserving edges.