An adaptive algorithm for fast moving object detection from dynamic background based on cobebook

In automatic videosurveillance system, the first step is the extraction of moving objects. In this work, we investigate a new approach for foreground-background segmentation by using a modified Codebook algorithm. This approach exploit the perceptual information to optimize the detection. It is an adaptive strategy which is proposed to to reduce the computational complexity of the foreground detection algorithm while maintaining its global accuracy. The algorithm uses superpixels which model the spatial dependencies between pixels and controls actively the distribution of these superpixels by focusing them on possible awaited location of foreground objects. The performance evaluation is performed on public sequences that propose sequences with a dynamic background and the experimental results demonstrate an increase in the detection speed.