吉林大学学报(理学版)

• 计算机科学 • 上一篇    下一篇

基于压缩感知和离散余弦变换的数字图像水印算法

张虹   

  1. 三峡大学 湖北省水电工程智能视觉监测重点实验室, 湖北 宜昌 443002
  • 收稿日期:2016-08-29 出版日期:2017-11-26 发布日期:2017-11-29
  • 通讯作者: 张虹 E-mail:licewhet@sina.com

Digital Image Watermarking Algorithm Based onCompressive Sensing and Discrete Cosine Transform

ZHANG Hong   

  1. Hubei Key Laboratory of Intelligent Vision Based Monitoring for Hydroelectric Engineering, China Three Gorges University, Yichang 443002, Hubei Province, China
  • Received:2016-08-29 Online:2017-11-26 Published:2017-11-29
  • Contact: ZHANG Hong E-mail:licewhet@sina.com

摘要: 为了使水印信息更好地隐藏于原始图像中, 加快水印信息检测, 设计一种基于压缩感知和离散余弦变换的数字图像水印算法. 首先将原始载体图像划分成多个子块, 采用离散余弦变换对各子块进行分解; 然后采用压缩感知对水印信息进行处理, 减少嵌入的信息量; 最后将水印信息嵌入到原始载体子块的相应位置, 并通过匹配跟踪算法实现水印信息的检测. 从主观和客观两方面对数字图像水印算法的性能进行评价, 结果表明, 该算法的水印具有良好的不可感知性, 对抵抗常见攻击具有很好的鲁棒性, 解决了水印透明性与鲁棒性之间的矛盾, 且加快了水印提取速度.

关键词: 离散余波变换, 压缩传感理论, 数字图像水印, 匹配跟踪算法, 水印检测

Abstract: In order to hide watermark information in the original image better and accelerate detection of the watermark information, the author designed a digital image watermarking algorithm based on compressive sensing and discrete cosine transform (DCT). Firstly, original carrier image was divided into several sub blocks, and discrete cosine transform was used to decompose each sub block. Secondly, compressed sensing was used to process watermark information
 to reduce the amount of embedded information. Finally, the watermark information was embedded into the corresponding position of the original carrier sub block, and the watermark detection was realized by matching pursuit algorithm. The performance of digital image watermarking algorithm was evaluated from two aspects of subjective and objective. The results show that the watermark of this algorithm has good perception, and has good robustness against common attacks. It solves contradiction between watermark transparency and robustness, and accelerates the speed of watermark extraction.

Key words: digital image watermarking, discrete cosine transform, watermark detection, compressive sensing theory, matching pursuit algorithm

中图分类号: 

  • TP391