Professional paper

Txt and Tif File Compression Using Lzw, Huffman, and Arithmetic Coding

Deni Klen orcid.org/0009-0008-7473-8653 ; University of Rijeka, Faculty of Engineering, Vukovarska 58, 51000 Rijeka, Croatia *
Jonatan Lerga ; University of Rijeka, Faculty of Engineering, Vukovarska 58, 51000 Rijeka, Croatia
Iena Petrijevčanin ; University of Rijeka, Center for Artificial Intelligence and Cybersecurity, R. Matejčić 2, 51000 Rijeka, Croatia

* Corresponding author.

Abstract

The paper compares compression methods such as Lempel-Ziv-Welch (LZW), Huffman, and arithmetic
coding applied to different large text and image datasets. Comparison is done based on metrics such
as execution time and compression ratio. LZW produced results of about 30 % median compression
ratio for all text records and a median of about 70 % for image records. In addition, Huffman coding
produced a compression rate of about 40 % median for text data and a median of about 55 % for
image data. Finally, arithmetic coding yielded results of about 70 % median for text compression and
about 55 % median for image data compression. The time required was lowest for LZW, followed by
Huffman, and worst for arithmetic coding.

Keywords

coding, LZW, Huffman, arithmetic coding, compression

Hrčak ID:

317229

URI

https://hrcak.srce.hr/317229

Publication date:

16.5.2024.

Visits: 241 *