Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Peripheral devices for the right light

: Weber, Werner; James, Lex; Velzen, Arjan van; Vulcano, Bruno; Stalpers, Micha; Stadt, Arend van de; Döge, Jens; Rabaud, Wilfried

Volltext (PDF; )

LED professional Review (2015), Nr.48, S.92-97
ISSN: 1993-890X
Bundesministerium für Bildung und Forschung BMBF
IKT 2020 - Forschung für Innovation; Enlight
Zeitschriftenaufsatz, Elektronische Publikation
Fraunhofer IIS, Institutsteil Entwurfsautomatisierung (EAS) ()

The closer the relation between the user and his requirements in terms of light provided in space and time, the better the comfort. These demands were fulfilled by advanced sensors and EnLight’s distributed rules based intelligence concept. Various physical parameters are important and need to be measured to allow proper decision-making such as occupancy, position of people in a room, level of outside lighting and temperature. Within the scope of the EnLight project, Fraunhofer IIS/EAS developed a low-power presence detector, based on an image-sensor system-on-chip (SoC). In contrast to traditional solutions, such as passive infrared (PIR) sensors, this system processes the image in the visible and near infrared spectrum, and derives texture features from visual information. A self-learning multi-modal algorithm analyzes the scene at a frame rate of up to ten frames per second. People can be localized within the field of view and distinguished from technical objects according to the active area and typical motion. In contrast to gray-scale algorithms the chosen approach is highly tolerant to brightness variations occurring due to incoming sunlight or switched luminaires.