基于分块DCT的自适应扩频水印算法  

Adaptive Watermarking Algorithm Based on Block-DCT and Spread Spectrum

作　　者：饶智坚[1] 常建平[1] 

机构地区：[1]南京航空航天大学信息科学与技术学院,南京210016

出　　处：《科技广场》2005年第5期9-14,共6页Science Mosaic

摘　　要：本文提出了一个利用块分类的DCT域自适应扩频图像水印算法,嵌人的水印是一个可视二值图像,用两个不相关的伪随机序列分别代表水印信息中的0和1,从而达到扩频的目的,将扩频后的水印信号,嵌入到分块DCT域的中低频段系数中。在嵌人过程中充分考虑到了局部图像的复杂度,能够自适应地调整嵌人强度,提取水印信号无需使用原始图像。仿真实验表明:图像经过JPEG有损压缩、低通与中值滤波、剪切等图像处理操作后仍是稳健的。An adaptive watermarking technique based on block-DCT and spread spectrum is presented in this paper. The embed watermark is a visible binary image, We use two independent pseudo-random sequence substitute the ‘0’and‘1’ in the binary image, thus achieve the goal of spread spectrum. The modulated signal is embedded in the low-middle frequency band of block- DCT, The scheme can adaptive adjust embedding strength according to local image complexity and can extract the watermark without the original image .The simulation results demonstrate that the method is robust to most of the image procession operations such as lossy JPEG compression, low pass filtering, median filtering, cropping etc.

关 键 词：分块DCT 扩频 HVS 

分 类 号：TP309.7[自动化与计算机技术—计算机系统结构]