通信学报 ›› 2013, Vol. 34 ›› Issue (3): 148-156.doi: 10.3969/j.issn.1000-436x.2013.03.019

• 技术报告 • 上一篇    下一篇

基于子块区域分割和自嵌入技术的全盲多功能图像水印算法

叶天语   

  1. 浙江工商大学 信息与电子工程学院,浙江 杭州310018
  • 出版日期:2013-03-25 发布日期:2017-07-20
  • 基金资助:
    浙江省教育厅科研基金资助项目

Perfectly blind image watermarking scheme with multi-purpose based on region segment for sub-block and self-embedding technology

Tian-yu YE   

  1. College of Information & Electronic Engineering,Zhejiang Gongshang University, Hangzhou 310018,China
  • Online:2013-03-25 Published:2017-07-20
  • Supported by:
    The Scientific Research Fund of Zhejiang Provincial Education Department

摘要:

利用子块区域分割思想和自嵌入技术提出了一种只利用攻击图像就能进行版权鉴别和内容认证的全盲多功能图像水印算法。将原始图像分割成不重叠的子块,对每个子块进行离散余弦变换,将每个子块离散余弦变换矩阵分成为区域1和区域2两部分,然后对区域1进行奇异值分解,通过判断奇异值均值的最高位数字奇偶性产生特征水印,然后通过调整区域2的2个离散余弦变换交流系数大小自嵌入特征水印,最后对每个子块进行逆离散余弦变换得到含水印图像。算法通过结合将区域1产生的特征水印自嵌入区域2和在区域2盲提取认证水印实现全盲检测和多功能。实验结果表明,算法既能实现版权保护又能实现内容认证,而且还具有区分恶意篡改和无意篡改的能力。

关键词: 数字水印, 多功能水印, 全盲检测, 区域分割, 自嵌入技术

Abstract:

A perfectly blind image watermarking scheme with multi-purpose was proposed through region segment for sub-block and self-embedding technology, which could achieve copyright identification and content authentication only with the help of an attacked image. The original image was split into non-overlapping blocks, and each block was con-ducted with discreet cosine transformation. Each block's discreet cosine transformation matrix was divided into region 1 and region 2. Then, after the region 1 was conducted with singular value decomposition,a feature watermark was derived from judging the parity of the first digit from average of singular values. Moreover, the feature watermark was self-embedded into region 2 by adjusting two discreet cosine transformation alternative current coefficients. Finally, a watermarked image was obtained after conducting inverse discrete cosine transformation on each block. The proposed algorithm achieved perfectly blind detection and multi-purpose by combining self-embedding feature watermark from re-gion 1 into region 2 and blindly extracting authentication watermark n region 2. Experimental results show that the pro-posed algorithm can achieve both copyright protection and content authentication, and has the ability to distinguish mali-cious tamper and unintentional tamper.

Key words: digital watermarking, multi-purpose watermark, perfectly blind detection, region segment, self-embedding technology

No Suggested Reading articles found!