Data compression is a key problem in the applications of hyperspectral images. To meet the demand

based on 2-D wavelet transform

this paper proposes a lossy compression method using band grouping and vector quantization. Every band of hyperspectral images is decomposed by biorthogonal filters and the coefficients are coded by Kronecker-Product Gain-Shape Vector Quantization. The experimental results have shown that the computational complexity is reduced. The high compression ratio is achieved and the spectral characteristics are preserved.