Robust Real Time Moving People Detection in Surveillance Scenarios

In this paper an improved real time algorithm for detectingpedestrians in surveillance video is proposed. Thealgorithm is based on people appearance and defines a personmodel as the union of four models of body parts. Firstly,motion segmentation is performed to detect moving pixels.Then, moving regions are extracted and tracked. Finally,the detected moving objects are classified as human or nonhumanobjects. In order to test and validate the algorithm,we have developed a dataset containing annotated surveillancesequences of different complexity levels focused onthe pedestrians detection. Experimental results over thisdataset show that our approach performs considerably wellat real time and even better than other real and non-realtime approaches from the state of art.