Clustering-Based Image Sparse Denoising in Wireless Multimedia Sensor Networks

With the increasing interest in the deployment of wireless multimedia sensor networks (WMSN), new challenges have arisen with the complexity and high noise level of the monitoring environment. Given that the noise severely impairs the quality and visibility of video images perceived by sensors, video image denoising naturally becomes the key to ensure the validity and reliability of the WMSN video monitoring. In this paper, the sparse denoising algorithm via clustering-based sparse representation is proposed. Firstly, WMSN images are, respectively, clustered based on the pixel intensity of regions of interest (ROIs), which are determined in terms of Bayesian theorem. Secondly, in the light of nonlocal self-similarity regularizer provided by the ROI-based WMSN images clustering, clustering-based sparse representation builds a new sparse denoising model exploiting both sparsity and nonlocal self-similarity to improve the quality of reconstructed images. At last, a surrogate-function-based iterative shrinkage solution has been developed to solve the double-header $$l_{1}$$l1-optimization problem. Experimental results showed that the performance of the approach to image denoising is competitive, qualitative, as well as quantitative, and suitable for the WMSN video image denoising.