Abstract:
This paper proposes a novel robust steganalysis system, adopts the genetic algorithm, embed is the watermark information into the DCT domain, and implements an optimized embedding of the watermark information. The genetic algorithm can keep good perceptual transparency, and don′t make distortion of extracted watermark. Experimental results show that the algorithm has the desired properties such as robustness against noise, but insufficiency against smoothing attack and sharping attack operation.
Key words:
digital watermark,
steganalysis,
steganography,
genetic algorithm,
fitness function
摘要: 提出一种新颖的、鲁棒的隐写系统,采用遗传算法将水印信息嵌入到图像的DCT变换域中,实现水印信息的最优化嵌入。该算法提取出的水印无失真,可以改善水印系统不可感知性和不可统计性。实验结果表明,该方法抵抗噪声较强,但其平滑攻击、锐化攻击的能力较差。
关键词:
数字水印,
隐写分析,
隐写术,
遗传算法,
适应度函数
CLC Number:
MAO Jia-fa; LIN Jia-jun; DAI Meng. Application of Genetic Algorithm for Embedding of Digital Watermark in Image DCT Domain[J]. Computer Engineering, 2008, 34(4): 167-169.
毛家发;林家骏;戴 蒙. 遗传算法在图像DCT域中嵌入数字水印的应用[J]. 计算机工程, 2008, 34(4): 167-169.