Efficient impulse noise reduction via local directional gradients and fuzzy logic

This paper proposes an efficient noise reduction method for image restoration. A novel pixel classification method by using fuzzy logic and local gradients is first introduced to discriminate noise and noise-free pixels from corrupted images. Then, a switching filtering scheme is employed to noisy images. The noise pixels will be filtered by a weighted median filter with adaptive weights when the noise-free pixels are preserved in the filtered images. Both numerical comparisons and perceptual observations are demonstrated in the experimental simulations. It is found that the proposed method achieves a fairly better performance than a number of other existing methods in noise reduction and image restoration. (C) 2008 Elsevier B.V. All rights reserved.