Computer Vision Laboratory


Field CT 


The goal of this US Government-funded project is to develop and demonstrate a low cost, field deployable, stationary 3D x-ray computed tomography (CT) system that will image root system architectures in situ. The proposed system is based on UHV’s unique low cost linear x-ray tube technology and sophisticated image reconstruction and segmentation algorithms developed at the University of Massachusetts, Lowell and the University of Nottingham. Once the CT images have been segmented to distinguish roots from soil, a quantitative 3D representation of root architecture will be expressed in Root System Markup Language (RSML). The prototype system will be evaluated in cornfields with two types of soil at the University of Wisconsin and Texas A&M University.
Segmentation of root material from soil in x-ray CT images is a challenging computer vision problem. The project will therefore leverage both recent developments in deep machine learning and the segmentation by tracking approach that has proved successful within the University’s Hounsfield Facility.


Funding Information

ARPA-e CT is supported by a grant from ARPA-e , the research arm of the US Department of Energy

Project Team

UHV Technologies (Kentucky)

University of Massachusetts (Lowell)

University of Wisconsin

Texas A&M University

CVL (Tony Pridmore, Stefan Mairhofer, Reza Soltaninejad)

CVL Contact

For more informaiton, please contact Tony Pridmore

Computer Vision Laboratory

The University of Nottingham
Jubilee Campus
Wollaton Road
Nottingham, NG8 1BB

telephone: +44 (0) 115 8466543