Evaluation of Huffman and Arithmetic Algorithms for Multimedia\n Compression Standards

Compression is a technique to reduce the quantity of data without excessively\nreducing the quality of the multimedia data. The transition and storing of\ncompressed multimedia data is much faster and more efficient than original\nuncompressed multimedia data. There are various techniques and standards for\nmultimedia data compression, especially for image compression such as the JPEG\nand JPEG2000 standards. These standards consist of different functions such as\ncolor space conversion and entropy coding. Arithmetic and Huffman coding are\nnormally used in the entropy coding phase. In this paper we try to answer the\nfollowing question. Which entropy coding, arithmetic or Huffman, is more\nsuitable compared to other from the compression ratio, performance, and\nimplementation points of view? We have implemented and tested Huffman and\narithmetic algorithms. Our implemented results show that compression ratio of\narithmetic coding is better than Huffman coding, while the performance of the\nHuffman coding is higher than Arithmetic coding. In addition, implementation of\nHuffman coding is much easier than the Arithmetic coding.\n