Compression of hyperspectral data by spatial/spectral discrete cosine transform
P. Baizert, Mark R. Pickering, Michael J. Ryan · 2002
The preferred vector quantization technique for the lossless compression of hyperspectral data has been demonstrated to be a combination of mean-normalised vector quantization (M-NVQ) in the spatial domain and the discrete cosine transform (DCT) techniques in the spectral domain. Results for a spatial M-NVQ/spectral DCT coder are between 1.5 and 2.5 times better than the compression ratios obtained by the M-NVQ technique alone. This work shows that, for low distortion levels, further improvements in compression ratio (approximately 0.5 bits/pixel) can be obtained by replacing the spatial M-NVQ technique with the two-dimensional DCT.