The objective of this research is the development of a mathematical framework that enables the identification, characterization and matching of patterns in imaging data with certain guarantees. The application areas for this project include medical imaging, autonomous driving systems, and segmentation of natural and biological images. This work has been partially funded by the NSF under award CNS-1239323.
Mobile sensor networks are often composed of agents with weak processing capabilities and some means of mobility. However, recent developments in embedded systems have enabled more powerful and portable processing units capable of analyzing complex data streams in real time. Systems with such capabilities are well-suited for environmental monitoring using a combination of cameras, microphones, and sensors for temperature, air-quality, and pressure. Still there are few compact platforms that combine state of the art hardware with accessible software, an open source design, and an affordable price. The WolfBot platform offers a balance between capabilities, accessibility, cost and an open-design.