摘要
针对图像水印的透明性和鲁棒性的矛盾问题,以及图像水印在图像压缩攻击中鲁棒性较差的问题,提出基于LWT-DCT-SVD混合域的抗压缩鲁棒性图像水印方案.该方案应用混沌序列和置乱变换加密水印,提高水印的安全性,并选取LWT-DCT变换的低频系数进行SVD变换,并在奇异值中采用自适应量化嵌入水印,实现盲检测.仿真实验结果表明,水印保持较好的透明性,同时在噪声污染、低通滤波、图像缩放、恶意剪切等常见图像处理具有较强的鲁棒性,尤其对于图像压缩攻击,本水印方案显示出较强的抗压缩特征.
The work aims to propose an anti-compression robust image watermarking scheme based on LWTDCT-SVD hybrid domain to solve the problems of the contradiction between the invisibility and robustness of image watermarking and the poor robustness of image watermarking in image compression attacks.In this scheme,the watermark is encrypted by chaotic sequence and permutation transform to improve the security of the watermark,and the low frequency coefficients of LWT-DCT transform are selected for SVD transform,and the watermark is embedded in the singular value by adaptive quantization to implement blind detection.The simulation results show that the watermark is transparent and robust in common image processing such as noise pollution,low-pass filtering,image scaling and malicious cutting,especially for image compression attack.This watermarking scheme shows strong anti-compression characteristics.
作者
杨志疆
YANG Zhijiang(School of Electronic Information,Zhangzhou Institute of Technology,Zhangzhou,Fujian 363000,China)
出处
《肇庆学院学报》
2025年第2期69-74,共6页
Journal of Zhaoqing University
基金
福建省教育厅中青年教师教育科研项目(JAT220685)
漳州职业技术学院课题(ZZY2021B130)
漳州职业技术学院课题(ZZYKYK24005)。
关键词
提升小波变换
离散余弦变换
奇异值分解
抗压缩
鲁棒性
lift wavelet transform(LWT)
discrete consine transform(DCT)
singular value decomposition(SVD)
anti-compression
robustness
