Title: A maximum a-posteriori identification criterion for wavelet domain watermarking

Authors: Tek Ming Ng, Hari Krishna Garg

Addresses: Faculty of Engineering, Department of Electrical and Computer Engineering, National University of Singapore, 4 Engineering Drive 3, 117576, Singapore. ' Faculty of Engineering, Department of Electrical and Computer Engineering, National University of Singapore, 4 Engineering Drive 3, 117576, Singapore

Abstract: A detector based on Maximum A-posteriori Probability (MAP) criterion has been introduced for identifying image watermark in the Discrete Cosine Transform (DCT) domain. This type of detector has been shown to be optimum and robust to common image processing operations. In this paper, an MAP detector in the Discrete Wavelet Transform (DWT) domain is proposed. It is based on modelling the wavelet coefficients by a generalised Gaussian distribution. Simulations are used to compare the proposed detector with the conventional correlation detector in terms of detection effectiveness.

Keywords: MAP detector; correlation detector; wavelet transform; generalised Gaussian; identification criterion; maximum a-posteriori probability; image watermarking; discrete cosine transform; image processing; modelling; simulation.

DOI: 10.1504/IJWMC.2009.029346

International Journal of Wireless and Mobile Computing, 2009 Vol.3 No.4, pp.265 - 270

Published online: 22 Nov 2009 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article