Lossless interframe motion-compensated coding schemes for biomedical image sequences

A new lossless coding scheme is proposed, especially adapted to optimize the storage capacities of dynamic medical imagery without any numerical distorsion after coding-decoding procedures. It basically uses a lossy (i.e an adaptive predictive one) plus lossless residual encoding scheme ([LieS9], [Rab91]). The predictive coder is based on motion compensation techniques. The essential originalities of this algorithmic work is, firstly, to propose and compare different motion models (from purely trauslational to affine models.) and, then, to select the optimal lossless encoding technique related to the p. d. f of residual motion compensated errors. Finally we propose comparative tests on real coronarography image sequences.

[1]  G. Furlan An enhancement to universal modeling algorithm context for real-time applications to image compression , 1991, [Proceedings] ICASSP 91: 1991 International Conference on Acoustics, Speech, and Signal Processing.

[2]  Jorma Rissanen,et al.  A multiplication-free multialphabet arithmetic code , 1989, IEEE Trans. Commun..

[3]  D. Walker,et al.  Improved Pel-Recursive Motion Compensation , 1984, IEEE Trans. Commun..

[4]  Donald E. Knuth,et al.  Dynamic Huffman Coding , 1985, J. Algorithms.

[5]  Henri Nicolas,et al.  Compact motion representation based on global features for semantic image sequence coding , 1991, Other Conferences.

[6]  J Liénard Real-time distortionless high-factor compression scheme. , 1989, Medical physics.

[7]  JORMA RISSANEN,et al.  A universal data compression system , 1983, IEEE Trans. Inf. Theory.