The wealth of spatio-spectral data obtained using modern hyperspectral imager has brought in new challenges in the analysis and extraction of useful information from hyperspectral datasets. That is why we are developing an on-board processing system which will reduce the data downlinked to our ground stations. A significant part of the on-board pipeline is target detection. The objective of target detection algorithms is to find an object of interest in the hyperspectral image. For example, a target detection system can determine the algae species based on its spectral signature and many specific plant pigments as well as certain organic and inorganic compounds.

Detecting targets with different dimensionality reduction techniques.

Written by: Đorđije Bošković