A real-time algorithm for error recovery in remote video-based surveillance applications

This work presents a real-time post-processing error-recovery algorithm explicitly devoted at enhancing the performances of outdoor video-surveillance systems working in remote modality. The aim of the proposed algorithm is to distinguish between changed blocks due to variations in the observed scene and noise-altered blocks that contain errors caused by channel noise. Such errors can be corrected by directly exploiting the considerable spatio-temporal redundancy of the encoded digital source without using any additional information. Experimental results, obtained through colour JPEG transmission simulations performed in the context of an actual remote video-surveillance system, compare the proposed scheme with different concealment schemes. It is proven that substantial improvements both in terms of perceptual quality and performance of the overall video-surveillance system are possible, by using the proposed algorithm as a stand-alone module or in conjunction with a relatively low-re...