Title: Adaptive background subtraction using fuzzy logic

Authors: M. Sivabalakrishnan, D. Manjula

Addresses: Department of CSE, College of Engineering, Anna University, Chennai, 600 025, India. ' Department of CSE, College of Engineering, Anna University, Chennai, 600 025, India

Abstract: Visual surveillance applications such as object identification, movement tracking, and activity monitoring require reliable moving object detection as an initial processing step. The process that segments moving objects from the stationary scene is termed background subtraction. Extracting a background from an image is the enabling step for many high level vision processing tasks such as object tracking and activity analysis. In this paper, a novel fuzzy approach is used for background subtraction with a particular interest to the problem of silhouette detection. Experimental results demonstrated that fuzzy system is efficient. Features are extracted from image regions, accumulated the feature information over time, fused high-level knowledge with low-level features, and built a time-varying background model. A problem with the system is that by adapting the background model, objects moved are difficult to handle. The test results show the feasibility of the proposed algorithm.

Keywords: median filters; background subtraction; foreground detection; fuzzy background subtraction; SAD; segmentation; fuzzy logic; visual surveillance; moving objects; object detection; image processing; object tracking; activity analysis.

DOI: 10.1504/IJMIS.2010.039239

International Journal of Multimedia Intelligence and Security, 2010 Vol.1 No.4, pp.392 - 401

Published online: 28 Mar 2011 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article