Image compression in the wavelet domain using an AR texture model with compressed initial conditions

Authors:
M. H. Perng;H. H. Lin
Affiliations:
Department of Power Mechanical Engineering, National Tsing Hua University, Hsinchu, Taiwan, R. O. C.;Department of Power Mechanical Engineering, National Tsing Hua University, Hsinchu, Taiwan, R. O. C.
Venue:
ISPRA'06 Proceedings of the 5th WSEAS International Conference on Signal Processing, Robotics and Automation
Year:
2006

Citing 8
Cited 0

An evaluation of stochastic models for analysis and synthesis of gray-scale texture

Pattern Recognition Letters
Choice of a 2-D causal autoregressive texture model using information criteria

Pattern Recognition Letters
The JPEG2000 still image coding system: an overview

IEEE Transactions on Consumer Electronics
Image compression by texture modeling in the wavelet domain

IEEE Transactions on Image Processing
A new approach for subset 2-D AR model identification for describing textures

IEEE Transactions on Image Processing
Visually improved image compression by combining a conventional wavelet-codec with texture modeling

IEEE Transactions on Image Processing
Improved MPEG-4 still texture image coding under noisy environment

IEEE Transactions on Image Processing
A scalable MPEG-4 wavelet-based visual texture compression system with optimized memory organization

IEEE Transactions on Circuits and Systems for Video Technology

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper present a texture compression technique for still images based on the wavelet transform and the auto-regressive (AR) texture model in order to increase the compression ratio with a minimal loss of image quality. First the influences of the initial condition and the order of an AR model on the resulting texture model are investigated to serve as a theoretical foundation for the proposed approach. To further the compression ratio, this paper also presents a texture compressing technique using an auto-regressive texture model with compressed initial conditions. Results show that the AR model is better than a random texture model when the order of the AR model is adequately chosen, and compression of the initial conditions in the AR model can significantly improve the compression ratio without a noticeable loss of image quality.