Compressive hyperspectral imaging mask optimization.

Binbin Lv,Jiamin Wu,Chenggang Yan,Xinhong Hao,Guiguang Ding,Yongdong Zhang,Qionghai Dai
DOI: https://doi.org/10.1145/3240876.3240921
2018-01-01
Abstract:Hyperspectral imaging is a hot topic nowadays. It is an urgent problem to be solved how to achieve swift hyperspectral imaging. In this thesis, our primary purpose is to further optimize how to place a mask in front of a sensor in order to achieve compressed Hyperspectral imaging. We apply optimized projection matrix, matrix differential, projection analysis and other related knowledge to optimizing this realistic matter. After we simply introduce the background of hyperspectral imaging, we will firstly present the basic principle of compressed hyperspectral imaging based on mask, and then mainly analyze the way to achieve projection matrix optimizing algorithm and the challenges these sort of realistic problems face. Finally, we compare the experiment results of these two methods, and the rebuilding results before and after the optimizing.
What problem does this paper attempt to address?