一种基于多小波的盲的数字图象水印技术

A Multiwavelet-Based Blind Watermarking for Digital Image

以一个图案作水印且提取时不需要原图象的数字水印技术 (blind waterm arking)面临许多挑战 ,因而 ,这种技术的研究文献很少 .本文首次提出了一种基于多小波 (multiwavelet)变换的盲的数字图象水印技术 .与标量小波不同 ,在图象的多小波变换系数中 ,同一尺度且同一方向上存在两个 (或以上 )对应不同小波的细节子图 ,本文中的方法正是利用多小波变换这一独特性质嵌入一幅二值图案 .为了增加水印的稳健性 ,我们采用 (7,4 )汉明编码 (Hamming code)将水印的校验图也嵌入到多小波变换域中 .特别是 ,水印图案的提取不需要原始图象 .实验结果表明 ,这种方法对JPEG压缩、平滑滤波、锐化、噪声、对比度增强、重量化。

The watermarking in which the watermark is a visually recognizable pattern and watermark recovery does not need original image (blind watermarking) remains the most challenging problem , so the research on this technique is rare in the literature by now. This paper has first proposed a multiwavelet based blind watermarking for digital image. Unlike the scalar wavelet case, there are two different 'detail' coefficient blocks in the same direction and in the same resolution. According to this unique character , we embed a binary pattern into the multiwavlet domain of an image. To be more robustness , we use (7,4) Hamming code to encode the watermark and also embed its parity pattern into the multiwavlet domain of an image . Especially, the watermark recovery does not require the original image. Experimental results show that the proposed technique has good PSNR between watermarked image and original image and high robustness to common image processing operation and the JPEG lossy compression.