A new motion detection algorithm based on Σ– Δ background estimation

• Published in: Pattern recognition letters Vol. 28; no. 3; pp. 320 - 328
• Main Authors: Manzanera, Antoine, Richefeu, Julien C.
• Format: Journal Article Conference Proceeding
• Language: English
• Published: Amsterdam Elsevier B.V 01.02.2007; Elsevier
• Subjects: Applied sciences; Background estimation; Computer Science; Computer Vision and Pattern Recognition; Detection, estimation, filtering, equalization, prediction; Exact sciences and technology; Image Processing; Information, signal and communications theory; Miscellaneous; Motion detection; Recursive filtering; Signal and communications theory; Signal processing; Signal, noise; Telecommunications and information theory; Background estimation; Motion detection; Recursive filtering; Motion estimation; Image processing; Recursive method; Signal processing; Motion detection: Background estimation; Algorithm; Non linear operator
• ISSN: 0167-8655; 1872-7344; 1872-7344
• DOI: 10.1016/j.patrec.2006.04.007

Motion detection using a stationary camera can be done by estimating the static scene (background). In that purpose, we propose a new method based on a simple recursive non linear operator, the Σ– Δ filter. Used along with a spatiotemporal regularization algorithm, it allows robust, computationally efficient and accurate motion detection. To deal with complex scenes containing a wide range of motion models with very different time constants, we propose a generalization of the basic model to multiple Σ– Δ estimation.