Please use this identifier to cite or link to this item: https://dspace.univ-ouargla.dz/jspui/handle/123456789/2493
Title: A Novel Digital Watermarking Method Based on Multi-Level DWT and New Class of Reciprocal-Orthogonal Parametric Transforms ROPT
Authors: Ammar DAHMANI, Ahmed LOUCHENE,
Keywords: Digital watermarking
Gaussian noise
JPEG compression
ROPT
DWT
still image
data hiding
Issue Date: 19-Dec-2013
Abstract: since the second half of the 1990’ several works, dealing with image watermarking, have been proposed. The majority of those works are based on the use of transformed (frequency) domain provided by various transforms namely DCT, DFT, DWT…etc. Further, several hybrid techniques have been proposed, among those based on block DCT/DFT and those relying on DWT. through the work presented in this paper, authors propose a new hybrid watermarking method for still and grey scale images. It combines two transforms, namely the multi-level discreet wavelet transform (DWT) decomposition and a new class of reciprocal- orthogonal parametric transforms (ROPT). The basic idea behind the proposed method is to use a new method of generating keys by exploiting the huge number of independent parameters (3N/2) provided by the ROPT as supplement watermarking key added to the one used for inserting the watermark. To get rid of the disadvantage of the ROPT, which gives a non uniform partition of frequencies, the low-pass (LL) sub-band obtained by DWT is used. By this way we ensure on the one hand that the region frequencies in where the watermark will be inserted are all from the same range, on the other hand to ensure the robustness of the watermark against some attacks. Experimental results demonstrate that the proposed method is robust against JPEG compression, low pass filtering, added noise and the combination of these attacks.
Description: The INTERNATIONAL CONFERENCE ON ELECTRONICS & OIL: FROM THEORY TO APPLICATIONS March 05-06, 2013, Ouargla, Algeria
URI: http://hdl.handle.net/123456789/2493
ISSN: waf
Appears in Collections:3. Faculté des sciences appliquées

Files in This Item:
File Description SizeFormat 
Ammar_DAHMANI.pdf194,36 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.