A novel coding algorithm for video surveillance

In this paper, a new coding algorithm designed for video surveillance is proposed. By use of the techniques such as context switch, inertia-based motion estimation, early-out DCT transform, region-based quantization and so on, this algorithm provides some attractive features for video surveillance. The developed algorithm supports multiple video channels, has error-resilience and alarming functions, and leads to high compression ratio. It is also sensitive to moving objects and feasible for browsing. Simulation results show that the proposed algorithm is more suitable than MPEG-1 for the application of video surveillance.