Search

An adaptive nonlocal-means (ANL-means) algorithm for image denoising is proposed in this work. It employs the singular value decomposition (SVD) method and the K-means clustering (K-means) technique to achieve robust block classification in noisy images. Then, a local window is adaptively adjusted to match the local property of a block and a rotated matching algorithm that aligns the dominant orientation of a local region is adopted for similarity matching. Furthermore, the noise level is estimated using the block classification result and the Laplacian operator. Experimental results are given to demonstrate the superior denoising performance of the proposed ANL-means denoising technique over various image denoising benchmarks in terms of the PSNR value and perceptual quality comparison, where images corrupted by additive white Gaussian noise (AWGN) are tested.

Two challenging situations for video frame rate up-conversion (FRUC) are first identified and analyzed; namely, when the input video has abrupt illumination change and/or a low frame rate. Then, a low-complexity processing technique and robust FRUC algorithm are proposed to address these two issues. The proposed algorithm utilizes a translational motion vector model of the first- and the second-order and detects the continuity of these motion vectors. Additionally, in order to improve perceptual quality of interpolated frame, spatial smoothness criterion is employed. The superior performance of the proposed algorithm has been tested extensively and representative examples are given in this work.

Low-complexity error concealment techniques for mobile applications are studied extensively in this paper. The boundary matching algorithm (BMA) is an attractive choice for mobile video concealment due to its low complexity. Here, we examine a variant of BMA called the outer boundary matching algorithm (OBMA). Although BMA and OBMA are similar in their design principle, it is empirically observed that OBMA outperforms BMA by a significant margin (typically, 0.5 dB or higher) while maintaining the same level of complexity. In this work, we attempt to explain the superior performance of OBMA, and conclude that OBMA provides an excellent tradeoff between the complexity and the quality of concealed video for a wide range of test video sequences and error conditions.