Entropy-Constrained Multiresolution Vector Quantisation for Image Coding

Article type: Research Article

Authors: Cieplinski, Leszek

Affiliations: Centre for Vision, Speech and Signal Processing, School of Electronic Engineering, Information Technology and Mathematics, University of Surrey, Guildford, United Kingdom. [email protected]

Note: [] Address for correspondence: Visual Information Laboratory, Mitsubishi Electric Information Technology Centre Europe, 20 Frederick Sanger Road, Guildford, Surrey GU2 5YD, UK

Abstract: A new image coding scheme based on multiresolution image representation and entropy-constrained vector quantisation is proposed. The proposed technique is evaluated using a series of images. The results show a performance improvement over the previously proposed multiresolution scheme at relatively high bitrates. The reasons for worse performance at low bitrates are discussed.

Keywords: image compression, multiresolution image coding, vector quantization

DOI: 10.3233/FI-1998-34403

Journal: Fundamenta Informaticae, vol. 34, no. 4, pp. 389-396, 1998