Improving digital watermarking fidelity using fast neural network for adaptive wavelet synthesis
MetadanePokaż pełny rekord
This paper introduces a new adaptive algorithm for digital watermark embedding in wavelet domain. The proposed algorithm performs adaptive mother wavelet synthesis based on a low frequency component energy maximization. The algorithm is based on an orthogonal neural network. We demonstrate that the presented adaptive method can improve both the correlation between an extracted watermark and an embedded watermark, as well as the fidelity of an image. The proposed algorithm is applied to improve well known wavelet based embedding algorithms.
- Artykuły (WFTIiMS)