WeebleVideo – wide angle field-of-view video sensor networks

Pushkar Sambhoos, Ahmad Bilal Hasan, Richard Han, Tom Lookabaugh, Jane Mulligan

Research output: Chapter in Book/Report/Conference proceedingConference proceeding contributionpeer-review


Low cost networks of wireless sensors can be distributed to provide information about an environment. Even a network of sensors providing scalar measurements (for instance, of temperature) presents both formidable challenges in terms of integrating and interpreting measurements over space and time, and important opportunities in extended observations. Cameras are particularly powerful multidimensional sensors for dispersing in unknown environments for surveillance and tracking of activity. Understanding the spatial patterns of such activity requires the camera network to self-organize in terms of understanding relative positions of nodes. Cameras also pose problems for resource limited motes because of the high volumes of image data for local processing or transmission. We describe a self-righting or weeble node architecture for camera networks based on integrating a low cost camera into the Mica2 sensor node platform. The node uses a wide field of view lens (typically called a fish eye lens) which allows us to capture a very broad region around the node providing greater view overlap between the nodes and generally a larger frame for identifying and tracking activity.
Original languageEnglish
Title of host publicationWorking Notes of the International Workshop on Distributed Smart Cameras (DSC-06)
EditorsBernhard Rinner, Wayne Wolf
Number of pages5
Publication statusPublished - 2006
Externally publishedYes
EventSenSys'06: 4th International Conference on Embedded Networked Sensor Systems - Boulder, CO, United States
Duration: 31 Oct 20063 Nov 2006


ConferenceSenSys'06: 4th International Conference on Embedded Networked Sensor Systems
Country/TerritoryUnited States
CityBoulder, CO


Dive into the research topics of 'WeebleVideo – wide angle field-of-view video sensor networks'. Together they form a unique fingerprint.

Cite this