摘要
针对分块自嵌入水印算法的安全性和不能有效判断图像篡改块的问题,提出一种基于混沌置乱的分块自嵌入算法。该算法将由图像块生成的水印按空间位置生成二值水印图像,利用混沌序列对其置乱加密后嵌入到图像的最低位,认证时根据设定的阈值能有效定位图像内容被篡改的图像块。实验仿真和理论分析表明,该算法在保持原算法篡改恢复效果的基础上,无论是否有水印被改变,均能有效判断图像内容被篡改的图像块,且具有更高的安全性。
A chaos-based scramble self-embedding watermarking algorithm was proposed to locate tamper block on the watermarked images and improve security of the existing self-embedding fragile watermarking algorithms that could not effectively locate tamper block. In proposed algorithm, the binary watermark image was generated from the watermark blocks putted together according to their positions, and the improved security binary watermark scrambled by chaotic systems was embedded into the LSB of the image data. While image authentication, the block of tampered content could be effectively localized according to the predefined threshold. Theoretic analysis and simulation results show that the proposed algorithms may not only improve security, but it also effectively locates tamper blocks whether watermark is altered or not.
出处
《通信学报》
EI
CSCD
北大核心
2006年第7期80-86,93,共8页
Journal on Communications
基金
教育部新世纪优秀人才支持计划基金资助项目(NECT-05-0794)
四川省青年基金资助项目(03ZQ026-033)
国家预研基金资助项目(51430804QT2201)~~
关键词
脆弱水印
自嵌入
混沌
置乱
fragile watermarking
self-embedding
chaos
scramble
