The Use of Quadtree Range Domain Partitioning with Fast Double Moment Descriptors to Enhance FIC of Colored Image

Sultan, Bushra and George, Loay and Hassan, Nidaa (2018) The Use of Quadtree Range Domain Partitioning with Fast Double Moment Descriptors to Enhance FIC of Colored Image. ARO-The Scientific Journal of Koya University, 6 (1). pp. 13-22. ISSN 24109355

[img]
Preview
Text
ARO-10207-VOL6.No1.2018.ISSUE10-PP13-22.pdf

Download (1MB) | Preview
Official URL: http://aro.koyauniversity.org/editor/submission/20...

Abstract

In this paper, an enhanced fractal image compression system (FIC) is proposed; it is based on using both symmetry prediction and blocks indexing to speed up the blocks matching process. The proposed FIC uses quad tree as variable range block partitioning mechanism. two criteria’s for guiding the partitioning decision are used: The first one uses sobel-based edge magnitude, whereas the second uses the contrast of block. A new set of moment descriptors are introduced, they differ from the previously used descriptors by their ability to emphasize the weights of different parts of each block. The effectiveness of all possible combinations of double moments descriptors has been investigated. Furthermore, a fast computation mechanism is introduced to compute the moments attended to improve the overall computation cost. the results of applied tests on the system for the cases “variable and fixed range” block partitioning mechanism indicated that the variable partitioning scheme can produce better results than fixed partitioning one (that is, 4 × 4 block) in term of compression ratio, faster than and PSNR does not significantly decreased.

Item Type: Article
Uncontrolled Keywords: Fractal image compression, Iterated function system, Moments features, Quadtree
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions: Aro-The Scientific Journal of Koya University > VOL 6, NO 1 (2018)
Depositing User: Dr Salah Ismaeel Yahya
Date Deposited: 29 Apr 2018 08:06
Last Modified: 29 Apr 2018 08:06
URI: http://eprints.koyauniversity.org/id/eprint/144

Actions (login required)

View Item View Item