Occupancy Pattern Recognition with Infrared Array Sensors: A Bayesian Approach to Multi-body Tracking

Abstract

Thermal vision systems based on low-cost IR array sensors are becoming attractive in many smart living scenarios. This paper proposes a Bayesian framework for recognition and discrimination of body motions based on real-time analysis of thermal signatures. Unlike conventional frame-based methods, the proposed approach exploits a statistical model for the extraction of body-induced thermal signatures and a mobility model for tracking multi-body motions inside an indoor area. This approach prevents typical detection problems and can be also used in presence of interfering thermal sources such as heaters, radiators and other thermal devices. The Bayesian method is verified experimentally for ceiling mounted sensors and shows high accuracy and robustness even in cases where thermal signatures are closer to the ambient temperature.