Holly Rushmeier
Yale University, CT·Computer Science
A regular camera records color in a simple way suited to human eyes. Hyperspectral imaging (HSI) instead records light precisely across many wavelengths, from ultraviolet to infrared.
- Fields
Computer Science
Engineering
- Topics
hyperspectral imaging
spectroscopy
computer vision
- Grants
- 1 known grant, through Mar 2027
What they work on
- Establish an open-source hyperspectral imaging ecosystem, built around a tool and data hub, that accepts continuous contributions from instrument makers, researchers, software providers and users.
- Create a comprehensive hyperspectral image data resource, with protocols and software standards for sharing data across machine learning and artificial intelligence applications.
- Release well-documented open-source software tools and hardware designs that make hyperspectral imaging easier for new adopters to use.
More about the research
This reveals details people cannot see, which is useful in biology, medicine, drug making, materials, cultural heritage and space operations. Wider use has been held back by scattered patents, isolated and immature software and data, steep learning curves and missing industry standards. This project addresses those barriers.
Why it matters
Shared, open hyperspectral imaging tools and data could speed up work on natural resource monitoring, climate change assessment, artifact authentication, cancer diagnostics, pathogen detection and quality control for pharmaceuticals and seeds. A common data hub could also support machine learning uses of the technology and lower the barrier for new users.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Stuart Chalk
University of North Florida, FL·Computer Science
Develop core components of KnowLedger, an open digital research notebook that records data and notes, connects to online resources, and can be tailored to different scientific workflows.
Nian Zhang
University of the District of Columbia, DC·Computer Science
Develop a data preprocessing method for infrared backscatter hyperspectral images that handles class imbalance, where threat-chemical pixels are rare, and unlabeled data.
Noah Snavely
Cornell University, NY·Computer Science
Develop compositional implicit neural scene representations that pair a network's ability to capture fine detail with physical models of scene structure, such as light fields.
The school
Yale University: 73 labs on PI Pages, $1.5B a year on research, 438 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.