基于混合优化算法的图像盲水印技术  

Image Blind Watermark Based on Hybrid Optimization Algorithm

作　　者：张国梅 邹华兴[2] 吴迪 ZHANG Guomei;ZOU Huaxing;WU Di(School of Computer Science and Engineering,Guangzhou Institute of Science and Technology,Guangzhou 510540,Guangdong,China;School of Information Engineering,Nanchang University,Nanchang 330031,Jiangxi,China;College of Computer Science and Technology,Harbin Engineering University,Harbin 150001,Heilongjiang,China)

机构地区：[1]广州理工学院计算机科学与工程学院,广东广州510540 [2]南昌大学信息工程学院,江西南昌330031 [3]哈尔滨工程大学计算机科学与技术学院,黑龙江哈尔滨150001

出　　处：《济南大学学报（自然科学版）》2022年第3期278-285,共8页Journal of University of Jinan(Science and Technology)

基　　金：国家自然科学基金项目(61601189);广东省普通高校科研项目(2019KTSCX243,2020KTSCX210)。

摘　　要：为了平衡水印系统约束与净负荷容量,提出一种结合量子粒子群优化与细菌觅食优化的混合优化算法,将提出的混合优化算法应用于图像盲水印;通过离散小波变换对测试图像的像素进行变换,选择测试图像的水平低频垂直高频和水平高频垂直低频子带,基于边缘熵和视觉熵的视觉特性,在子带中选择合适区块以提高水印图像质量;对选定的区块系数执行Schur变换,并隐藏感兴趣区域和患者电子记录水印;利用提出的混合优化算法选择最优阈值,将多个水印隐藏到测试图像中。结果表明,基于混合优化算法的图像盲水印技术性能优秀,可以抵御常见攻击,同时具有较好的鲁棒性和隐蔽性。To balance constraints of watermarking systems and net load capacity,a hybrid optimization algorithm com-bining quantum particle swarm optimization and bacterial foraging optimization was proposed and applied to image blind watermark.Discrete wavelet transform was used to transform pixels of test images,and horizontal low-fnequenly vertical high-frequency as well as horizontal high-frequency vertical low-frequency subbands of the test images were selected.Based on visual characteristics of edge entropy and visual entropy,appropriate blocks were selected in the subbands to improve watermark image quality.Schur transform was performed on selected block coefficients,and regions of interest and patient electronic record watermarks were hidden.The proposed hybrid optimization algorithm was used to select the optimal threshold to hide multiple watermarks in the test images.The results show that the image blind watermark based on the hybrid optimization algorithm has excellent performances,which can resist common attacks,and has good robustness and concealment.

关 键 词：混合优化算法 盲水印 感兴趣区域 鲁棒性 隐蔽性 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]