An Efficient Image Compression Using Bit Allocation based on Psychovisual Threshold

Ernawan, Ferda and Zuriani, Mustaffa and Bayuaji, Luhur (2016) An Efficient Image Compression Using Bit Allocation based on Psychovisual Threshold. Information (Japan), 19 (9B). pp. 4177-4182. ISSN 1343-4500. (Published)

[img]
Preview
PDF
visiogame_manuscript.pdf

Download (216kB) | Preview

Abstract

One of the main part of image compression is a quantization process which give a significant effect to the compression performance. However, image compression based on the quantization produces blocking effect or artifact image. This research proposes a novel bit allocation strategy which assigning an optimal budget of bits in image compression. The bit allocation is proposed to replace the role of the quantization process in image compression. The principle of psychovisual threshold is adopted to develop bit allocation strategy in the image compression. This quantitative research measures the optimal bit of image signals and manages the image quality level. The experimental results show the efficiency of the proposed bit allocation strategy, and that the proposed bit allocation can achieve the almost same compression rate performance while can significantly produces high quality image texture. When compared to JPEG compression, the image compression using bit allocation achieves bit rate savings of up to 4%. The quality image output provides minimum errors of artifact image. The quality image reconstruction improvement is up to 14% and the error reconstruction is reduced by up to 37%. Key Words: Bit allocation, Quantization table, Psychovisual threshold, Image compression

Item Type: Article
Additional Information: The 3rd International Conference on Computer Graphics, Visualization, Computer Vision and Game Technology 2016
Uncontrolled Keywords: Bit allocation, Quantization table, Psychovisual threshold, Image compression
Subjects: Q Science > Q Science (General)
T Technology > T Technology (General)
Faculty/Division: Faculty of Computer System And Software Engineering
Depositing User: Dr. Luhur Bayuaji
Date Deposited: 21 Sep 2016 02:27
Last Modified: 26 Feb 2018 01:56
URI: http://umpir.ump.edu.my/id/eprint/13071
Download Statistic: View Download Statistics

Actions (login required)

View Item View Item