Vehicle detection segmentation based on adaboost and Grabcut

Segmentation of moving regions in image sequences is a fundamental step in many vision systems including automated visual surveillance, human-machine interface. In this paper, we combine background subtraction algorithm with adaboost classifier to obtain exact moving areas. Grabcut segmentation is then used to further accurately segment the moving target areas. Our experiments show that our algorithm can achieve high reliability target detection with low false positive rate in complex situations.