Low Cost Wireless Image Sensor Networks for Visual Surveillance and Intrusion Detection Applications
Résumé
Wireless Image Sensor Networks (WISN) where sensor nodes are equipped with miniaturized visual CMOS cameras to provide visual information is a promising technology for situation-awareness, search&rescue applications or intrusion detection applications. In this paper, we present an off-the-shelf image sensor based on an Arduino Due board with a CMOS uCamII camera. The image sensor works with raw 128x128 image and implements a simple intrusion detection mechanism based on simple-differencing technique. The total time for getting the image, encoding it, detecting intrusion and transmitting it can be less than 2.3s. We detail the performance measures and using real hardware constraints, we also show how more accurate simulation model can be built for large-scale multi-hop networked configurations.