压缩感知在数字图像水印中的应用分析

Application of compressive sensing in digital image watermark

压缩感知是信号处理领域的新兴理论,首先阐述压缩感知的基本原则和理论框架,然后对基于正交小波变换下压缩感知的图像重构的效果进行比较分析。实验表明,压缩感知可以很好地运用于图像重构。将压缩感知理论与数字水印技术相结合提出一种基于压缩感知理论的RGB空间彩色图像水印算法。该算法充分利用压缩感知的稀疏性及压缩比的可调节性,控制水印信息的嵌入容量的同时很好地提高了水印嵌入的安全性。实验表明,对于一些常见的攻击,该算法具备很好的鲁棒性。

Compressive sensing is a new theory in the field of signal processing. the basic principles and theoretical framework of compressive sensing are briefly expounded in this paper. The results of the image reconstruction of compressed sensing based on the orthogonal wavelet transform are compared and analyzed. Experiment results show that the compressive sensing can be well applied to image reconstruction. A watermark algorithm for RGB space color images based on compressive sensing theory is proposed in combination with the compressive sensing theory and digital watermark technology. The algorithm makes full use of the sparsity of the compressive sensing and regularity of compression ratio to control the watermark information embedding capacity and improve the security of watermark embedding. The experiment results show that the algorithm has good robustness for some common attacks.