Reversible Watermarking Based on Prediction Error Expansion and Pixel Selection on Color Image
G.S.Raman1, C.Surya2, R. Balaji Ganesh3
1G.S.Raman, Associate Professor Department of Information Technology, KLN College of Information Technology, Sivangangai, India.
2C.Surya, PG Scholar, Department of Information Technology, KLN College of Information Technology, Sivagangai, India.
3R.Balaji Ganesh, CITE Research Scholar, MS University Thirunelveli, India.
Manuscript received on March 18, 2013. | Revised Manuscript received on April 10, 2013. | Manuscript published on April 30, 2013. | PP: 925-929 | Volume-2, Issue-4, April 2013. | Retrieval Number: D1500042413/2013©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Reversible watermarking enables embedding of valuable information in a host signal with no loss of host information. The conventional PEE exploit the similar inherent in the neighborhood of pixel that the difference expansion scheme. In our proposed system, the PEE technique is further investigated and an resourceful reversible watermarking scheme is deduced by incorporating in PEE two new techniques, namely, adaptive embedding and pixel selection. PEE technique embeds data consistently, using Embedded Zerotree Wavelet (EZW), Bitplan Complexity Segmentation (BPCS) based embedding is applied to embed on natural images. This avoids expanding pixels with huge prediction errors likewise it also reduces embedding impact by diminishing the maximum modification to pixel values. We as well put forward to selecting pixels of smooth area for data embedding and leave the rough pixels unchanged. In our method a more penetratingly effective method for data embedding and a better visual quality of watermarked image is observed.
Keywords: Pixel selection, Prediction error expansion, EZW, BPCS, Reversible watermarking.