Abstract:
Integer-coefficient Discrete Wavelet Transformation
(DWT) filters widely used in the literature are implemented
and investigated as spectral decorrelator for on-board
lossless hyperspectral image compression. As the
performance of spectral decorrelation step has direct impact
on the compression ratio (CR), it is important to employ the
most convenient spectral decorrelator in terms of low
computational complexity and high CR. Extensive tests
using AVIRIS image data set are carried out and CRs
corresponding to various subband decomposition levels are
presented within a lossless hyperspectral compression
framework. Results suggest that Cohen-DaubechiesFeauveau (CDF) 9/7 integer-coefficient wavelet transform
with five levels of spectral subband decomposition would be
an efficient spectral decorrelator for on-board lossless
hyperspectral image compression.