A fast lossless compression algorithm for Arabic textual images

In recent years, an unparalleled volume of textual information was transported over the Internet via email, chatting, blogging, twittering, digital libraries, and information retrieval systems. As the volume of text data has exceeded 40% of the total volume of traffic on the Internet, compressing textual data becomes imperative. Many algorithms were introduced and employed for this purpose including Huffman encoding, arithmetic encoding, the Ziv-Lempel family, Dynamic Markov Compression, and Burrow-Wheeler Transform. In this paper, a novel algorithm for compressing textual images is presented. The algorithm comprises of two parts: (i) a fixed-to-variable codebook; and (ii) row and column reduction coding scheme, RCRC. Simulation results on a large number of Arabic textual images show that this algorithm has a compression ratio of approximately 87%, which exceeds published results including those of JBIG2.