Compression Techniques are Lovable or Hateful: For Discrete Tone Images

Abstract

The amount of data transmission especially images over internet are rapidly increasing day by day. The effective lossless image compression becomes a greater challenge now than ever. The efficiently compressed images can be useful to remotely access multimedia file at a faster rate with lesser burden on network infrastructure. Though numerous researches has been done on compression of Continuous Tone (CT) images, only few has concentrated on Discrete Tone (DT) images. The nature of CT and DT images are contrast to each other. In this study, the existing image compression techniques are applied to DT images and the results are analyzed. To carry out his work, we have collected an own DT image dataset which consists of 11 reference images with its several distorted versions. The dataset contains a total of 71 images and the existing image compression techniques such as Lempel Ziv Markov chain Algorithm (LZMA), Prediction by Partial Matching (PPM), Burrows Wheeler Transform (BWT), Lempel Ziv Welch (LZW) coding, Deflate, LZ77 and Deflate64. The comparison results imply that PPM, LZW and Deflate64 achieve better compression than other methods. At the same time, Deflate and LZ77 achieves negative compression where the value of compression ratio crosses one.