Intelligent Video Surveillance with Abandoned Object Detection and Multiple Pedestrian Counting

We present a novel intelligent video surveillance system with efficient detection of abandoned objects and counting number of pedestrians. In the proposed algorithm the adaptively generated background enables to solve problems of illumination change and occlusions. After building the adaptive background model, the counting procedure starts to augment number of detected objects. Experimental results show that the proposed system outperforms existing abandoned object detection and pedestrian counting methods.