International Conference on Instrumentation, Measurement, Circuits and Systems (ICIMCS 2011)
137 A Novel Adaptive Watermarking Method and its Steganalysis Based on Spam Data Mining
Download citation file:
- Ris (Zotero)
- Reference Manager
In the age of the information, while people conveniently obtaining and exchanging the information, how to prove the ownership of information is become an important issue.
Watermark can be some words, signs and order numbers etc. and in the digital watermark, the loss of the embedded information is the same as the loss of the rights. Above all, digital watermark must have more powerful robustness and no perceptibility.
Thus, we based on Least Significant Bit (LSB) to propose the algorithm of high robustness and high transparency. This algorithm conducted repeatedly embedding on the high bit of the watermark image and used the algorithm of the LSB data hiding at the same time.
In the experiment, we used salt and pepper noise to test the robustness that we proposed, and used the steganalysis technology of Steganalysis by Subtractive Pixel Adjacency Matrix (SPAM), which is proposed by Fridrich et al. in 2010 to detect the no perceptibility. The method we proposed can be seen to effectively discriminate the watermark after being attacked in this experiment and it owns no perceptibility.