Addressing Data Management for Custom Built Hyperspectral Microscopy Station

článek ve sborníku ve WoS nebo Scopus

en

Hyperspectral (HS) imaging, originally developed for satellite applications as a remote sensing spatiospectral analysis technique, has been adopted by numerous scientific fields. By combining HS imaging and microscopy, we can capture not only spectral and spatial information but also non-destructively uncover compositional and chemical information, which is especially beneficial for applications in agriculture, food safety and inspection, medical research, microbiology and algae research. To assist in this endeavor, we have developed a modular pushbroom HS microscope platform that allows automated HS measurement to be conducted. This naturally poses a problem in the form of proper data management, as the volume and data storage requirements for such a type of system are essential to plan for in advance. In this paper, we address this problem and present our HS data management workflow with complete data and metadata collection, suitable format selection, and data compression, all with the FAIR data principles in mind.

Hyperspectral (HS) imaging, originally developed for satellite applications as a remote sensing spatiospectral analysis technique, has been adopted by numerous scientific fields. By combining HS imaging and microscopy, we can capture not only spectral and spatial information but also non-destructively uncover compositional and chemical information, which is especially beneficial for applications in agriculture, food safety and inspection, medical research, microbiology and algae research. To assist in this endeavor, we have developed a modular pushbroom HS microscope platform that allows automated HS measurement to be conducted. This naturally poses a problem in the form of proper data management, as the volume and data storage requirements for such a type of system are essential to plan for in advance. In this paper, we address this problem and present our HS data management workflow with complete data and metadata collection, suitable format selection, and data compression, all with the FAIR data principles in mind.

data management; dimensionality reduction; image compression; machine learning; pushbroom hyperspectral imaging

19.02.2025

IEEE Computer Society

Helsinki, Finland

979-8-3315-1313-9

2024 14th Workshop on Hyperspectral Imaging and Signal Processing: Evolution in Remote Sensing (WHISPERS)

5