Real-time change detection methods for video-surveillance systems with mobile camera

In this paper, a video-surveillance system based on a mobile-camera is presented. The proposed real-time method is able to detect an object in a video sequence from a non-static camera. In particular during an off-line phase the system creates a panoramic multi-layer background image. In the on-line phase the system compares the acquired images with a portion of the panoramic background. Different change detection methods are analyzed. Experimental results are presented in order to validate the proposed methods. The proposed method could be used for extending efficient algorithms for scene understanding already developed and tested for fixed cameras to a mobile camera environment.