摘要
提出一种基于视觉掩蔽的小波域二维数字水印新算法.该算法先将水印和图像进行多尺度分解,将分解的水印低频系数嵌入到具有相同尺寸的图像的低频系数中,重构得到水印图像,而水印分解中的高频系数将被保留,在水印检测过程中作为密钥使用.为了实现视觉掩蔽,在嵌入过程中,利用图像小波分解人眼的视觉系统特性,使得水印在不可见和鲁棒性都得到提高.检测水印时需要原始的未加水印的图像.实验结果表明,算法对压缩、加噪、线性滤波、剪裁等具有较高的鲁棒性.
A new watermark algorithm based on visual masking in wavelet domain is proposed. Based on JPEG2000 standardization, the image and watermark are both transformed into the discrete wavelet domain. Then, the lower frequency coefficients of the watermark are embedded into those of the image with the multiresolution data fusion. Finally, the associated coefficients are restructured to obtain the embedded image. Especially, in this embedding process, the higher frequency coefficients of the watermark multiresolution data are reserved and will be used as the secret key in the detection process. In order to better fit visual masking, the behavior of the HVS to the watermarking problem in the image wavelet decomposition is used. So it can improve the invisibility and robustness. The original image is used in the proposed detection algorithm. The experimental results show that it is robust against compression, adding noise, filtering, and cropping.
出处
《西安电子科技大学学报》
EI
CAS
CSCD
北大核心
2003年第1期38-41,共4页
Journal of Xidian University
基金
陕西省自然科学基金资助项目(D7010308)
