Nonparametric model for background subtraction-英文文献.pdfVIP

Non-parametric Mo del for Background Subtraction Ahmed Elgammal, David Harwo o d, Larry Davis Computer Vision Lab oratory University of Maryland, College Park, MD 20742, USA felgammal,harwo o d,lsd g@ Abstract. Background subtraction is a metho d typically used to seg- ment moving regions in image sequences taken from a static camera by comparing each new frame to a mo del of the scene background. We present a novel non-parametric background mo del and a background subtraction approach. The mo del can handle situations where the back- ground of the scene is cluttered and not completely static but contains small motions such as tree branches and bushes. The mo del estimates the probability of observing pixel intensity values based on a sample of intensity values for each pixel. The mo del adapts quickly to changes in the scene which enables very sensitive detection of moving targets. We also show how the mo del can use color information to suppress detec- tion of shadows. The implementation of the mo del runs in real-time for b oth gray level and color imagery. Evaluation shows that this approach achieves very sensitive detection with very low false alarm rates. Key words: visual motion, active and real time vision, motion detection, non-parametric estimation, visual surveillance, shadow detection 1 Intro duction The detection of unusual motion is the rst stage in many automated visual surveillance applications. It is always desirable to achieve very high sensitivity in the detection of moving ob jects with the lowest p ossible false alarm rates. Background subtraction is a metho d typic