摘要
为了提高图像水印算法的鲁棒性,提出了一种DWT域中基于块分类的稳健性图像水印算法。首先定义了幻方变换矩阵,对水印图像按密钥进行分块,按水印图像的复杂程度确定分块的大小;对原始图像进行离散Haar小波变换,然后将置乱加密后的二值水印图像嵌入到Haar小波低频子带。基于块的二次幻方置乱形式直观、分散度较好、易于实现且具有较强的实用性和灵活性;该水印算法运算速度快,安全性高,并且对噪声、JPEG压缩、剪切、平滑滤波等图像失真处理均具有很好的鲁棒性。
In order to improve the robustness of image watermark algorithm,a robust image watermark algorithm based on block classification in DWT field was put forward.Magic square transformation matrix was firstly defined,watermark image is blocked according to cryptographic key,and the block size is decided by the intricacy of watermark image;discrete Haar wavelet transformation was performed to the original image,then the scrambling ciphered binary watermark image was embedded into the Haar wavelet low-frequency sub-band.The secondary magic square scrambling based on block is intuitive in style with relatively good divergence,it is easy to implement and strongly practical and flexible;the operational speed of the watermark algorithm is quick,its security is high and it has very good robustness to the processing of image distortion,such as noise,JPEG compression,clipping and smoothing filtering,etc.
出处
《包装工程》
CAS
CSCD
北大核心
2010年第19期102-105,共4页
Packaging Engineering
关键词
幻方置乱
DWT变换
鲁棒性
magic square scrambling
DWT transformation
robustness
