Lossy Compression of CFA Image Based on Multiwavelet Packet

A lossy compression scheme for Bayer images is proposed in this paper. Recently, it was found that compression-first schemes outperform the conventional demosaicking-first schemes in terms of output image quality. Balanced multiwavelet packet transforms effectively remove CFA image correlation between frequency bands. Wavelet coefficients shuffling exploring subband correlation makes it suitable for zerotree coding. Improved SPIHT algorithm further exploits data correlation in different direction under the same resolution using one symbol to denote three zerotrees while the SPIHT algorithm using three symbols. The simulation results show that: proposed compression scheme at low bit rate compared with JPEG2000 has better performance, and low complexity. CPSNR is 2 db higher than JPEG2000 at bit rate 1 and is 0.7 db higher than JPEG2000 at bit rate 2.