Real-time video analysis for intrusion detection in indoor environments

In this paper we propose a novel system for indoor video surveillance. Our system, starting from a sequence of images, is able to detect and track moving objects even in the presence of significant variations of scene illumination. After a first analysis and clustering of the luminance time changes, a classification algorithm based on a fuzzy logic approach is used to identify moving regions that really represent unexpected objects moving in the scene, while discarding reflections and luminance profile changes due to illumination variations. One key feature of our system is its modest computation complexity, which allows it to operate in real-time on a standard PC platform. The real-time implementation of the system has been tested on a wide variety of situations, proving its effectiveness and robustness.