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

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.

[1]  P. Bouthemy,et al.  Recovery of moving object masks in an image sequence using local spatiotemporal contextual information , 1993 .

[2]  Donald Geman,et al.  Stochastic relaxation, Gibbs distributions, and the Bayesian restoration of images , 1984 .

[3]  Anil K. Jain,et al.  Markov Random Field Texture Models , 1983, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[4]  John M. DeCatrel,et al.  Markov random fields on a SIMD machine for global region labelling , 1991, Defense, Security, and Sensing.

[5]  Daniela Dragomirescu,et al.  A cellular analog network for MRF-based video motion detection , 1999 .

[6]  Tomaso Poggio,et al.  Computational vision and regularization theory , 1985, Nature.

[7]  Jin Luo,et al.  Computing motion using analog and binary resistive networks , 1988, Computer.

[8]  Christophe DumontierR.  Étude et mise en œuvre temps réel d'un algorithme de détection de mouvement par approche markovienne , 1996 .

[9]  J. Besag.  On the Statistical Analysis of Dirty Pictures , 1986 .

[10]  Charles A. Bouman,et al.  A multiscale random field model for Bayesian image segmentation , 1994, IEEE Trans. Image Process..

[11]  David W. Murray,et al.  Scene Segmentation from Visual Motion Using Global Optimization , 1987, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[12]  F. Heitz,et al.  Real-time collision avoidance at road-crossings on board the Prometheus-ProLab 2 vehicle , 1994, Proceedings of the Intelligent Vehicles '94 Symposium.

[13]  Alice Caplier,et al.  Real-Time Implementations of an MRF-based Motion Detection Algorithm , 1998, Real Time Imaging.

[14]  P. Bouthemy,et al.  Detection of small and slow moving objects observed by a mobile camera , 1996 .

[15]  Tomaso Poggio,et al.  Computational vision and regularization theory , 1985, Nature.

[16]  Eric Dubois,et al.  Bayesian Estimation of Motion Vector Fields , 1992, IEEE Trans. Pattern Anal. Mach. Intell..

[17]  Franck Luthon,et al.  Real time implementation of an MRF-based motion detection algorithm on a DSP board , 1996, 1996 IEEE Digital Signal Processing Workshop Proceedings.