Recent technological advances created a new generation of hyperspectral sensors, be it for satellite remote sensing or in situ monitoring. The corresponding data open the door to numerous potential applications with a high societal impact (defense and security, biomedical imagery, monitoring of the environment etc.). The optimal exploitation of the information provided by these very high dimensional data (blessing of dimensionality) cannot be performed using traditional methods (curse of dimensionality). New specific developments and algorithms are required. This project focuses on an emerging topic, which is as challenging as it is promising : the analysis of hyperspectral video sequences. Spectral unmixing taking the variability of the endmembers into account and denoising will be the two key issues considered.
Monsieur jocelyn chanussot (Grenoble Images Paroles Signal Automatique)
The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.
GIPSA-Lab Grenoble Images Paroles Signal Automatique
Help of the ANR 221,999 euros
Beginning and duration of the scientific project: - 36 Months