Statistical modelling of full frame DCT coefficients

Author

Barni, M. ; Bartolini, F. ; Piva, A. ; Rigacci, F.

Author_Institution

Dipt. di Ing. Elettron., Univ. di Firenze, Florence, Italy

fYear

1998

fDate

8-11 Sept. 1998

Firstpage

1

Lastpage

4

Abstract

Because of the increasing influence it has in image watermarking applications, the estimation of the distribution shape of full frame DCT coefficients is here addressed. Based on previous analyses on block-DCT, the coefficients are first assumed to follow a Generalized Gaussian distribution. The shape parameter v is then evaluated according to the maximum likelihood criterion applied to a set of 170 natural images. The analysis has been further validated by using the χ² test-of-fit criterion. In contrast to the block-DCT case, experimental results prove that full frame DCT coefficients can be modeled without appreciable loss of performance by a Laplacian density function having variance decreasing with frequency.

Keywords

Gaussian distribution; discrete cosine transforms; image watermarking; maximum likelihood estimation; χ² test-of-fit criterion; Laplacian density function; block-DCT; distribution shape estimation; full-frame DCT coefficients; generalized Gaussian distribution; image watermarking application; maximum likelihood criterion; natural images; shape parameter; statistical modelling; Discrete cosine transforms; Laplace equations; Maximum likelihood estimation; Shape; Standards; Watermarking;

fLanguage

English

Publisher

ieee

Conference_Titel

Signal Processing Conference (EUSIPCO 1998), 9th European

Conference_Location

Rhodes

Print_ISBN

978-960-7620-06-4

Type

conf

Filename

7089901