Beyond stream processing — A distributed vision architecture for the Internet of Things

An `edge-based' image processing architecture for the internet of things (IoT) is devised, drawing on the existing image processing pipelines that are implemented in today's imaging modules in digital cameras and smartphones. A key element of this IoT framework is that image or video data does not have to be streamed across the network, but can be largely processed on the sensing nodes with metadata transmitted to intermediate control nodes that can detect certain events/conditions and trigger corresponding actions.

[1]  P. Corcoran,et al.  Biometric Access Control for Digital Media Streams in Home Networks , 2007, IEEE Transactions on Consumer Electronics.

[2]  Peter M. Corcoran,et al.  Automated sorting of consumer image collections using face and peripheral region image classifiers , 2005, IEEE Transactions on Consumer Electronics.

[3]  Paul A. Viola,et al.  Rapid object detection using a boosted cascade of simple features , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[4]  Peter Corcoran,et al.  Advanced hardware real time face detector , 2012, 2012 IEEE International Conference on Consumer Electronics (ICCE).