Keywords: reversible watermarking, video, motion estimation, prediction error, histogram modification
Received June 11, 2009; revised October 27, 2009; accepted February 9, 2010.
Communicated by H. Y. Mark Liao.
| Previous | [ 1] | [ 2] | [ 3] | [ 4] | [ 5] | [ 6] | [ 7] | [ 8] | [ 9] | [ 10] | [ 11] | [ 12] | [ 13] | [ 14] | [ 15] | [ 16] | [ 17] | [ 18] | [ 19] | [ 20] | [ 21] | [ 22] | [ 23] | [ 24] | [ 25] |
¡@
XIAO ZENG, ZHEN-YONG CHEN, MING CHEN AND ZHANG XIONG
School of Computer Science and Engineering
Beihang University
Beijing, 100191 China
In this paper, a novel reversible video watermarking scheme using motion estimation
and prediction error expansion is presented, which can embed a large amount of secret
data into videos with imperceptible modification. Different with other reversible watermarking
schemes based on prediction error expansion, motion estimation is employed to
explore the relationship of neighboring frames and work out prediction errors, which significantly
sharpen the distribution of prediction errors. Then histogram modification is
used to expand the prediction errors, wherein a prediction error is changed by 1 or left unchanged
for embedding one secret bit. Due to the slight modification of pixels, high video
quality is preserved. In the phases of motion estimation and histogram modification, a little
side information used for extracting and recovering are generated, which will be combined
with the secret bits and embedded into the video. Experimental results demonstrate
that the proposed scheme provides a much larger watermark capacity and a better quality
of watermarked video than those of as reported in other state-of-the-art literature.
Received June 11, 2009; revised October 27, 2009; accepted February 9, 2010.
Communicated by H. Y. Mark Liao.