The paper presents new method for compression of correlated image sets. It is known that Karhunen-Loeve (KL) transform is most optimal representation for such a purpose. In our paper we use recently suggested fast algorithm of KL basis construction for compression of correlated image ensembles. The approach is based on fact that every KL basis function give maximum possible contribution in every image and this contribution decreases most quickly among all possible bases. So, we lossy compress every KL basis function by lossy DCT coding with essentially different loss that depends on the functions' contribution in the images.
展开▼