Entropy-threshold method for best basis selection

The methods which were adapted for the best basis selection optimise the decomposition by minimising the criterion cost function. This usually does not produce optimal rate-distortion (R-D) behaviour of the complete transform coder. This paper presents an extension of the well-known entropy-based be...

Full description

Saved in:
Bibliographic Details
Published in:Image and vision computing 2001-05, Vol.19 (6), p.477-484
Main Authors: Banjanin, B, Gergic, B, Planinsic, P, Cucej, Z
Format: Article
Language:English
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The methods which were adapted for the best basis selection optimise the decomposition by minimising the criterion cost function. This usually does not produce optimal rate-distortion (R-D) behaviour of the complete transform coder. This paper presents an extension of the well-known entropy-based best basis selection method titled entropy-threshold best basis selection. Using the threshold method, the criterion used in the original entropy-based method is strengthened and eliminates the portions of decomposition that are not cost effective. As the experiments illustrate, this new method exhibits great improvement in both R-D performance and simplified decomposition complexity. copyright 2001 Elsevier Science B.V.
ISSN:0262-8856