A minimalistic approach to image processing for low-power sensor networks
Teresa H. Ko, Nina M. Berry, Ronald Kyker · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 2004
Wireless sensor networks allow detailed sensing of otherwise unknown and inaccessible environments. While it would be beneficial to include cameras in a wireless sensor network because images are so rich in information, the power cost of transmitting an image across the wireless network can dramatically shorten the lifespan of the sensors. This paper investigates various compression techniques and what the cost of these algorithms would be to the lifespan of the sensor nodes. We further describe a new paradigm for cameras and wireless networks. Rather than focusing on transmitting images across the network, we show how an image can be processed locally for key features using simple detectors. Contrasted with traditional event detection systems that trigger a an image capture, this enables a new class of sensors which uses a low power imaging sensor to detect a variety of visual cues.