Real-time DSP implementation for MRF-based video motion detection

This paper describes the real-time implementation of a simple and robust mo tion detection algorithm based on Markov random field (MRF) modeling. MRF-b ased algorithms often require a significant amount of computations. The int rinsic parallel property of MRF modeling has led most of implementations to ward parallel machines and neural networks, but none of these approaches of fers an efficient solution for real-world (i.e., industrial) applications. Here, an alternative implementation for the problem at hand is presented yi elding a complete, efficient and autonomous real-time system for motion det ection. This system is based on a hybrid architecture, associating pipeline modules with one asynchronous module to perform the whole process, from vi deo acquisition to moving object masks visualization. A board prototype is presented and a processing rate of 15 images/s is achieved, showing the val idity of the approach.