Homogeneity-based directional sigma filtering of video noise

Abstract

This paper proposes a real-time method for the reduction of white Gaussian video noise. The method achieves a maximum gain of 4.8 dB and is capable of preserving image content. It adapts window size, weighting and behavior to both image content and noise level in order to optimize the filtering. It starts by detecting the intensity-homogeneous direction from 8 different candidates. A variant of the Sigma filter is then applied directionally. The filtering is performed along homogeneous areas and not across edges. For noisy images, the filtering is increased automatically by using the two most homogeneous directions with a larger kernel size. The proposed filter achieves better image preservation by turning off gradually for less noisy images. It works well for both highly noisy and good-quality images without the introduction of speed or hardware implementation challenges.

Surveillance video denoising based on background modeling

“…Reference [2] cited numbers of image denoising algorithms, and the most typical are mean filters and median filters, which lead to many variations [3][4] [5]. The implementation of such algorithms is simple, but the effect of denoising is not satisfying.…”

Robust video denoising for better subjective evaluation

“…Many video denoising algorithms have been proposed in literature and they may be classified as spatial domain methods [3][4][5][6] and transform domain methods [7][8][9][10][11][12][13]. In realtime video encoding system, the computational complexity of the wavelet domain method is too high.…”