数字图像水印算法研究及DSP实现

发布时间：2018-08-28 08:04

【摘要】：近年来,互联网络的飞速发展和数字多媒体技术的广泛应用极大地方便了人们的学习和工作,丰富了人们的日常生活,但数字产品的版权保护也显得日益迫切。在这种背景下,数字水印技术作为一种多媒体数据的版权保护和内容认证的新技术,成为了目前学术研究的一大热点,具有重要的理论研究价值和广泛的应用前景。本文主要针对数字图像水印算法进行了深入研究并将算法在DSP硬件平台上进行了实现。本文首先简要介绍数字水印技术的基本理论、数字水印算法的基本框架以及其分类和评价标准,总结了几种常见的水印信号预处理方法和水印攻击类型；其次在充分研究数字水印技术的基本理论及发展趋势的基础上,找出现有主要算法存在的问题及难点所在,将非抽样Contourlet变换、尺度不变特征变换、奇异值分解以及局部特征区域等理论引入到数字图像水印技术中,提出两种算法提高图像水印算法抗几何攻击的鲁棒性,提升算法的水印容量,仿真实验结果表明,这两种算法对于噪声、滤波、压缩以及各类几何攻击具有较好的鲁棒性；最后针对数字图像水印产品的实现问题,如具有水印认证功能的数码照相机,具有防伪功能的水印打印机、水印检测器等,设计了一个基于TMS320DM642的数字图像水印硬件系统,对TMS320DM642的结构及集成开发环境CCS进行了简要介绍,给出上位机和目标机各模块的软硬件设计以及最终的实验结果。
[Abstract]:In recent years, the rapid development of Internet and the wide application of digital multimedia technology have greatly facilitated people's study and work, enriched people's daily life, but the copyright protection of digital products has become increasingly urgent. Under this background, as a new technology of copyright protection and content authentication of multimedia data, digital watermarking technology has become a hot topic of academic research at present, and has important theoretical research value and wide application prospect. In this paper, the digital image watermarking algorithm is studied and implemented on DSP hardware platform. Firstly, this paper briefly introduces the basic theory of digital watermarking technology, the basic frame of digital watermarking algorithm, its classification and evaluation standard, and summarizes several common watermarking signal preprocessing methods and watermark attack types. Secondly, on the basis of studying the basic theory and developing trend of digital watermarking technology, the problems and difficulties of the main algorithms are found, and the non-sampling Contourlet transform and the scale invariant feature transformation are used. The singular value decomposition and local feature region are introduced into the digital image watermarking technology. Two algorithms are proposed to improve the robustness of the image watermarking algorithm against geometric attacks, and the watermarking capacity of the algorithm is improved. The simulation results show that the proposed algorithm is robust to geometric attack. These two algorithms are robust to noise, filtering, compression and various geometric attacks. Finally, the implementation of digital image watermarking products, such as digital camera with watermark authentication function, is discussed. A digital image watermarking hardware system based on TMS320DM642 is designed, which has anti-counterfeiting function, such as watermark printer, watermark detector and so on. The structure of TMS320DM642 and the integrated development environment CCS are introduced briefly. The hardware and software design of each module of upper computer and target machine and the final experimental results are given.
【学位授予单位】：湖南师范大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TP309.7;TP368.1

【相似文献】