Title :
A new filtering method for RST invariant image watermarking
Author :
Liu, Yan ; Zhao, Jiying
Author_Institution :
Sch. of Inf. Technol. & Eng., Ottawa Univ., Ont., Canada
Abstract :
Based on log-polar mapping, this paper presents a new filtering method. We compare our new filtering method with the classical matched filter, phase-only filter, binary phase-only filter, amplitude-only filter, and inverse filter. We found that our new method is the only one that is robust against rotation, scaling, and translation (RST) transformation. We use the filtering method in our new rotation, scaling, and translation (RST) invariant digital image watermarking scheme, to rectify the watermark position. The watermarking scheme does not need the original image to extract watermark and avoids exhaustive search. The three-dimensional plots of cross-correlation functions of different filters are presented and discussed.
Keywords :
correlation methods; digital filters; image coding; watermarking; RST invariant image watermarking; RST transformation; cross-correlation functions; digital watermark; filtering method; image rotation; image scaling; image translation; log-polar mapping; watermark position rectification; Data mining; Digital filters; Digital images; Image reconstruction; Information filtering; Information filters; Matched filters; Optical fibers; Optical filters; Watermarking;
Conference_Titel :
Haptic, Audio and Visual Environments and Their Applications, 2003. HAVE 2003. Proceedings. The 2nd IEEE Internatioal Workshop on
Print_ISBN :
0-7803-8108-4
DOI :
10.1109/HAVE.2003.1244733