摘要
提出一种水印算法的新思路,可以自适应地确定水印嵌入的区域和强度。基于HVS的人眼掩蔽效应对图像亮度和纹理敏感度的量化指标进行计算,图像先后两次分块。依据评分选择最佳嵌入区域后对其进行单尺度Retinex分解,将其分解为用于评价亮度信息的照度分量和评价纹理信息的反射分量。根据评价结果,设置不同的水印嵌入强度分别嵌入图像小波域中频的各分块中。提取的水印通过数学形态学处理以提高其可辨识度。嵌入水印的不可见性得到最大化实现(PSNR=40.0),鲁棒性较佳,能抵御信号处理和几何攻击。能根据图像特征自适应分析水印的最佳嵌入位置,在保证隐蔽性的同时嵌入尽可能多的水印信息;水印的安全性也得以保障,是一种应用面广、适应性强的数字水印算法。
A new solution of watermark algorithm is put forward to adaptively select the embedding area and intensity of watermark.Quantitative indicators are calculated by image brightness and texture sensitivity of human eyes masking effect based on HVS,and a image is partitioned twice to select the best embed area by score.After selecting the best embedded region according to the score,the single-scale Retinex decomposition is performed,which is decomposed into the radiance component used to evaluate the brightness information and the reflection component of the evaluation texture information.The watermark embedding different intensity is embed into the blocks of middle frequency in wavelet domain.The extracted watermark is processed by mathematical morphology to improve its identification.Invisibility is fully realized for the embed watermark（PSNR=40.0）.And it has good robustness with resisting signal process and geometric attack.The algorithm can adapt the best embedding position of watermark according to the image feature.Much more watermark information is embed on the basis of maintaining concealment and security.It is a digital watermark algorithm with extensive application and high adaptation.
出处
《光学技术》
CAS
CSCD
北大核心
2017年第6期555-560,共6页
Optical Technique
基金
上海理工大学科技发展项目(16KJFZ017)
上海市教委科研创新重点项目(13ZZ111)
关键词
Retinex分解
自适应
全息加密
数学形态学
retinex decomposition
self-adaption
holographic encryption
mathematical morphology
