A maximum a posteriori identification criterion for wavelet domain watermarking

Author

Ming, Ng Tek ; Garg, Hari Krishna

Author_Institution

Dept. of Electr. & Comput. Eng., Nat. Univ. of Singapore, Singapore

fYear

2004

fDate

23-24 March 2004

Firstpage

178

Lastpage

183

Abstract

Detector based on maximum a posteriori probability (MAP) criterion has been introduced for identifying image watermark in the discrete cosine transform domain. This type of detector has been shown to be optimum and robust to common image processing operations. An MAP detector in the discrete wavelet transform domain is proposed. It is based on modeling the wavelet coefficients by a generalized Gaussian distribution. Numerical experiments are used to compare the effectiveness of the proposed detector with the conventional correlation detector.

Keywords

Gaussian distribution; discrete cosine transforms; discrete wavelet transforms; image coding; maximum likelihood estimation; watermarking; MAP detector; correlation detector; discrete cosine transform domain; discrete wavelet transform domain; generalized Gaussian distribution; image processing operations; image watermark identification; maximum a posteriori probability criterion; wavelet domain watermarking; Detectors; Discrete cosine transforms; Discrete transforms; Discrete wavelet transforms; Gaussian distribution; Image processing; Robustness; Transform coding; Watermarking; Wavelet domain;

fLanguage

English

Publisher

ieee

Conference_Titel

Distributed Computing Systems Workshops, 2004. Proceedings. 24th International Conference on

Print_ISBN

0-7695-2087-1

Type

conf

DOI

10.1109/ICDCSW.2004.1284028

Filename

1284028