ABSTRACT

Extracting a background from an image is the enabling step for many high-level vision processing tasks, such as object tracking and activity analysis. Although there are a number of object extraction algorithms proposed in the literature, most approaches work efficiently only in constrained environments where the background is relatively simple and static. We extracted features 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 our system is that by adapting the background model, objects moved are difficult to handle. In order to reinsert them into the background, we run the risk of cutting off part of the object. In this paper, we develop a fuzzy logic inference system to detach the moving object from the background. Our experimental results demonstrate that the fuzzy inference system is very efficient and robust.