Title :
Real-time DSP implementation for MRF-based video motion detection
Author :
Dumontier, Christophe ; Luthon, Franck ; Charras, Jean-Pierre
Author_Institution :
Signal & Image Lab., Nat. Polytech. Inst., Grenoble, France
fDate :
10/1/1999 12:00:00 AM
Abstract :
This paper describes the real time implementation of a simple and robust motion detection algorithm based on Markov random field (MRF) modeling, MRF-based algorithms often require a significant amount of computations. The intrinsic parallel property of MRF modeling has led most of implementations toward parallel machines and neural networks, but none of these approaches offers an efficient solution for real-world (i.e., industrial) applications. Here, an alternative implementation for the problem at hand is presented yielding a complete, efficient and autonomous real-time system for motion detection. This system is based on a hybrid architecture, associating pipeline modules with one asynchronous module to perform the whole process, from video acquisition to moving object masks visualization. A board prototype is presented and a processing rate of 15 images/s is achieved, showing the validity of the approach
Keywords :
Markov processes; digital signal processing chips; motion estimation; real-time systems; video signal processing; MRF-based video motion detection; Markov random field; asynchronous module; board prototype; hybrid architecture; motion detection algorithm; object masks visualization; parallel property; pipeline modules; processing rate; real-time DSP implementation; video acquisition; Digital signal processing; Markov random fields; Motion detection; Neural networks; Parallel machines; Pipelines; Prototypes; Real time systems; Robustness; Visualization;
Journal_Title :
Image Processing, IEEE Transactions on