Abstract
In this paper, we present a digital image watermarking scheme that uses both human visual system and statistical properties. The scheme places watermark in discrete cosine transform (DCT) domain and spreads the watermark effect on the entire image. We have compared our proposed scheme with two other popular schemes and shown that our proposed scheme yields better results in terms of transparency, robustness and maximal capacity requirements.
Original language | English |
---|---|
Pages (from-to) | 145-149 |
Number of pages | 5 |
Journal | Pattern Recognition Letters |
Volume | 21 |
Issue number | 2 |
DOIs | |
State | Published - Feb 2000 |
Keywords
- Bounded normal distribution
- Entropy masking
- Watermarking