Mahmoud R. El-Sakka, "Context-based Dictionary Image Compression", International Conference on Computing and Information Technology, ICCIT'2012, pp. 140-143, March 2012, Medina, Saudi Arabia.
Classifying to Sub-Dictionaries (CSD) is a lossless data compression algorithm which combines the well-known Lempel-Ziv-Welch (LZW) data compression scheme and the Prediction by Partial Matching (PPM) scheme. CSD, as well as LZW and PPM, are originally introduced to compress text (one dimensional data). In this paper, our experimental work demonstrates the need for an adaptive approach to allow CSD to better utilize the two dimensional nature of image data so that compression performance can be further improved.