Using genetic algorithms to find new color transformations for JPEG-2000 image compression

Authors:
Mohammed S. Al-Rawi;Abdel-Latif Abu-Dalhoum;Yousef Salah;Wesam Al-Mobaideen;Ansar Khoury
Affiliations:
Computer Science Department, King Abdullah II School for Information Technology, The University of Jordan, Jordan;Computer Science Department, King Abdullah II School for Information Technology, The University of Jordan, Jordan;Computer Science Department, King Abdullah II School for Information Technology, The University of Jordan, Jordan;Computer Science Department, King Abdullah II School for Information Technology, The University of Jordan, Jordan;Computer Science Department, King Abdullah II School for Information Technology, The University of Jordan, Jordan
Venue:
WSEAS Transactions on Computers
Year:
2008

Citing 5
Cited 0

Adaptation in natural and artificial systems

Adaptation in natural and artificial systems
JPEG2000: the upcoming still image compression standard

Pattern Recognition Letters
Evolving Compression Preprocessors With Genetic Programming

GECCO '02 Proceedings of the Genetic and Evolutionary Computation Conference
JPEG2000 Standard for Image Compression: Concepts, Algorithms and VLSI Architectures

JPEG2000 Standard for Image Compression: Concepts, Algorithms and VLSI Architectures
The JPEG2000 still image coding system: an overview

IEEE Transactions on Consumer Electronics

Quantified Score

Hi-index	0.00

Visualization

Abstract

Image compression techniques play an important role in multimedia applications. The JPEG-2000 which is based on the wavelet transform is a promising image compression technique expected to replace the current discrete cosine transform based compression known as JPEG. In this paper, genetic algorithms are used to optimize the coefficients of the RGB to YCbCr color transform used in JPEG-2000 and to find alternate color transforms. Color transformation in JPEG-2000 is an early phase of the compression process intended to improve the compression performance. The matrix elements are optimized using fitness functions based on the root mean square error between the original and the reconstructed image. The resultant color transformations revealed an enhancement in the JPEG-2000 codec by.