Address-event imagers for sensor networks

摘要: Although imaging is an information-rich sensing modality, the use of cameras in sensor networks very often prohibited by factors such as power, computation cost, storage, communication bandwidth and privacy. In this paper we consider information selective privacy-preserving address-event imagers for networks. Instead providing full images with a high degree redundancy, our efforts design these specialize on selecting handful features from scene outputting representation. present initial results modeling evaluating sensors context Using three different platforms that have developed, illustrate how to model build emulator using models. We also lightweight classification scheme computational advantages sensors. The concludes evaluation algorithm feasibility study COTS components emulate inside network.