Hyperspectral image compression based on Tucker Decomposition and Discrete Cosine Transform

In this paper, an efficient method for Hyperspectral image compression based on the Tucker Decomposition (TD) and the Three Dimensional Discrete Cosine Transform (3D-DCT) is proposed. The core idea behind our proposed technique is to apply TD to the 3D-DCT coefficients of Hyperspectral image in order to not only exploit redundancies between bands but also to use spatial correlations of every image band and therefore, as simulation results applied to real Hyperspectral images demonstrate, it leads to a remarkable compression ratio with improved quality.