摘要
提出一种基于量化的Contourlet水印算法。该算法首先对原始水印信息进行置乱,采用伪随机序列周期延扩的方法,生成扩频水印。然后对载体图像进行Contourlet变换,最后使用量化的方法,将产生的扩频水印嵌入到选定的Contourlet域系数中。实验结果表明,该算法具有很好的透明性,对一般的信号处理,例如,JPEG压缩、模糊和锐化等具有较好的鲁棒性。
This paper presents a blind contourlet domain watermarking algorithm based on quantization. Firstly , the scrambling encrypted watermark was spread spectrum coded using the method of pseudo-random periodic extension. Secondly, contourlet transform was applied to the cover image. Thirdly, the final extended watermark was embedded into selected contourlet coefficients using quantization. The simulation results show that the algorithm is imperceptible and is robust against common signal processing such as JPEG compression, blurring and sharpening.
出处
《计算机安全》
2009年第5期17-21,共5页
Network & Computer Security