Optimal linear prediction of ordered bands and lifting wavelet transform applied to lossless compression of MODIS image

The enormous volumes and valuable applications of MODIS (Moderate Resolution Imaging Spectroradiometer) multispectral images make it need to be compressed losslessly with effective method. To solve this problem, the optimal linear prediction and band ordering are adopted here to exploit abundant spectral redundancy, lifting wavelet transform and SPIHT algorithm are used to eliminate the spatial redundancy of MODIS data. The optimal inter-band prediction sequence is specified by band ordering, and except the first band, only residual error images of other bands need to be encoded after band prediction. To avoid information loss, the optimal linear predictor is improved by rounding the prediction coefficients into integers, and D5/3 lifting wavelet is used to implement integer-to-integer wavelet transform of the band image, which also accelerate the compression. Finally, some lossless compression methods such as "WinRAR", "3DSPIHT" are compared with this compression method,. and the experimental results show that our method has more compression capability.