Towards a sensor for detecting human presence and activity

Yannick Benezeth (1) , Hélène Laurent (2) , Bruno Emile (2) , Christophe Rosenberger (3)

1 Orange Labs [Caen]
2 PRISME - Laboratoire Pluridisciplinaire de Recherche en Ingénierie des Systèmes, Mécanique et Energétique
3 Equipe SAFE - Laboratoire GREYC - UMR6072

In this paper, we propose a vision-based system for human detection and tracking in indoor environment allowing to collect higher level information on people activity. The developed presence sensor based on video analysis, using a static camera is ¯rst of all presented. Composed of three main steps, the ¯rst one consists in change detection using a background model updated at di®erent levels to manage the most common variations of the environment. A moving objects tracking based on interest points tracking is then performed. The classi¯cation step ¯nally relies on the use of statistical tools and multiple classi¯ers for the whole body and for the upper-body. The validation protocol, de¯ned by the industrial partners involved in the CAPTHOM project focusing among other things on "Energy Management in Building", is then detailed. Three applications integrated into the CAPTHOM draft ¯nally illustrate how the developed system can also help in collecting useful information for the building management system: occupancy detection and people counting as well as activity characterization and 3D location extend to a wide variety of buildings technology research areas such as human-centered environmental control including heating adjustment and demand-controlled ventilation, but also security and energy e±cient buildings.

Mots clés: human detection; presence sensor; occupancy number detection; activity characterization; people localization

Domaines: Cryptographie et sécurité [cs.CR]

Yannick Benezeth, Hélène Laurent, Bruno Emile, Christophe Rosenberger. Towards a sensor for detecting human presence and activity. Energy and Buildings, 2011, 43, pp.305-314. ⟨hal-00991093⟩