Hyperspectral image, video compression using sparse tucker tensor decomposition

Hyperspectral image and videos provide rich spectral information content, which facilitates accurate classification, unmixing, temporal change detection, so on. However, with the rapid improvements in technology, data size has increased many folds. To properly handle enormous volume, efficient methods are required to compress data. This paper proposes a multi-way approach for compression of hyperspectral or video sequence. In this approach, differential representation is first obtained. case images, difference between consecutive bands obtained videos, frames computed. next step, sparse Tucker tensor decomposition performed core Finally, corresponding factor matrices truncated encoded obtain compressed version transmission. The method utilises structure hence can be extended videos. Experimental results on several real imply that proposed obtains better efficiency terms ratio, signal noise ratio.