Memory efficient image compressing using lifting scheme

The transforming of mapping image data from RGB to YUV color space serves to reduce the correlation between components and it is suitable for human visual system. The line-based wavelet transform completes the column wavelet decomposition after each line is received in accumulative mode and it can reduce the memory utilization. The improved lifting scheme is adopted instead of the original Mallat arithmetic in this paper. It fulfills the CDF9/7 wavelet transform and reduces the complexity of the arithmetic. In this paper these two methods are both used. Experiment shows that these methods can save the memory utilization and reduce die time of the arithmetic implementation.