POSE: Phase II: An Open Source Hyperspectral Imaging Ecosystem

Award ID: 2303328

Traditional photography produces images for human perception with a simple representation of color. Hyperspectral Imaging (HSI) expands imaging beyond reproduction for human perception to precise spectral radiance recording ranging from the ultraviolet to the infrared wavelengths. The enhanced discernment enabled by HSI is used to benefit an abundance of applications across biology, medicine, pharmaceuticals, materials, cultural heritage preservation, and space operations. However, the application of HSI in the past has been inhibited by dispersed intellectual property, siloed and immature software tools and data, steep learning curves, and lack of industry standards. This project is releasing the potential of HSI by developing an open-source HSI ecosystem (HSI-OSE) with contributions from HSI instrument manufactures, academic researchers, software providers, and general users. The projects novelties are a sustainable organizational structure for the HSI-OSE established around a tool and data hub that enables continuous, open, and asynchronous contributions from registered users in the community. The project's impacts are accelerating progress in application areas such as natural resource monitoring, climate change assessment, artifact authentication, cancer diagnostics, pathogen detection, and quality control in pharmaceuticals and seeds. The HSI-OSE advances knowledge in the field of hyperspectral imaging by promoting synergy among the community of manufacturers, researchers, and users with tools and data sharing, as well as feature requests. With a central hub for HSI data resources, it enables further applications of the technology through machine learning (ML) and artificial intelligence (AI). Large amounts of data are generated in each HSI recording session depicting the attributes, properties, or characteristics of the imaging subject. Sophisticated software tools are required to process and interpret these data. This project creates a comprehensive HSI data resource and develops the protocols and standards for software development to enable data sharing across many machine ML/AI applications. The well-documented open-source software tools and hardware designs from the HSI-OSE also facilitate the entry of new adopters of the HSI technology. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Program Director: Florence Rabanal