Center for Embedded Network Sensing

The performance of camera sensors placed in harsh outdoor environments varies drastically because of factors such as lighting conditions, background contrast, luminance, etc. which are very hard to model. We present a data-driven approach to minimize the number of cameras while ensuring complete sensing coverage. A network of Cyclops cameras were deployed in the UCLA botanical gardens to detect moving people, Based on the data, we estimated detection ranges at a large number of points and found optimal camera locations.