The least statistically-dependent basis and its applications

Author

Saito, Naoki

Author_Institution

Dept. of Math., California Univ., Davis, CA, USA

Volume

1

fYear

1998

fDate

1-4 Nov. 1998

Firstpage

732

Abstract

Statistical independence is one of the most desirable properties of a coordinate system for representing and modeling images. In this paper we propose an algorithm to rapidly construct a coordinate system "closest" to the statistically independent one from a dictionary of bases such as the wavelet packets and local Fourier bases. The criterion is to minimize the sum of the coordinate-wise differential entropy and is quite different from the joint best basis (JBB) of Wickerhauser (1994). We demonstrate the use of the LSDB for image approximation and modeling, and compare its performance with Karhunen-Loeve basis (KLB) and JBB.

Keywords

Fourier transforms; approximation theory; image representation; minimum entropy methods; statistical analysis; wavelet transforms; Karhunen-Loeve basis; algorithm; approximation; coordinate system; dictionary; differential entropy minimisation; image approximation; image modeling; image representation; joint best basis; least statistically-dependent basis; local Fourier bases; probabilistic modelling; statistical independence; wavelet packets; Character generation; Dictionaries; Face; Humans; Image coding; Image generation; Mathematics; Principal component analysis; Quantization; Wavelet packets;

fLanguage

English

Publisher

ieee

Conference_Titel

Signals, Systems & Computers, 1998. Conference Record of the Thirty-Second Asilomar Conference on

Conference_Location

Pacific Grove, CA, USA

ISSN

1058-6393

Print_ISBN

0-7803-5148-7

Type

conf

DOI

10.1109/ACSSC.1998.750958

Filename

750958