1 January 2004 Compound document compression with model-based biased reconstruction
Author Affiliations +
J. of Electronic Imaging, 13(1), (2004). doi:10.1117/1.1631317
The usefulness of electronic document delivery and archives rests in large part on advances in compression technology. Documents can contain complex layouts with different data types, such as text and images, having different statistical characteristics. To achieve better image quality, it is important to make use of such characteristics in compression. We exploit the transform coefficient distributions for text and images. We show that the scheme in baseline JPEG does not lead to minimum mean-square error if we have models of these coefficients. Instead, we discuss an algorithm designed for this performance that involves first classifying the blocks, and then estimating the parameters to enable a biased reconstruction in the decompression value. Simulation results are shown to validate the advantages of this method.
Edmund Yin-Mun Lam, "Compound document compression with model-based biased reconstruction," Journal of Electronic Imaging 13(1), (1 January 2004). http://dx.doi.org/10.1117/1.1631317

Image compression


Signal to noise ratio

Image quality

Mathematical modeling

Model-based design

Stochastic processes


Efficient coding of residual images
Proceedings of SPIE (October 22 1993)
Color AC plasma panel barrier measurement system
Proceedings of SPIE (October 31 1996)
Efficient construction of saliency map
Proceedings of SPIE (February 10 2009)
Object-oriented reasoning in cognitive systems
Proceedings of SPIE (June 23 2000)

Back to Top