This paper describes a system for efficient storage and network transmission of images of historical documents. The images are segmented into two classes (paper and ink) with the use of an entropy-based segmentation algorithm. The information is then re-assembled, synthetising an image visually close to the original document. The synthesis process works in both classes, using information acquired in the segmentation phase. All the information needed to re-build the final synthetic image occupies, in average, 2 Kbytes performing a very efficient compression scheme.