基于奇偶量化的Contourlet变换域指纹图像水印算法

Fingerprint image watermarking algorithm using the quantization of parity based on Contourlet transform

提出了一种基于奇偶量化的Contourlet变换域指纹图像水印算法。原始指纹图像经过Contourlet变换后,分解为一系列多尺度、局部化、方向性的子带图像,选择在低频子带中嵌入经过二维Arnold置乱加密后的水印。嵌入水印时,采用奇偶量化算法修改低频子带的系数。水印提取时,不需要原始图像,实现了盲提取。实验结果表明,提出的指纹图像水印算法能够较好地抵抗JPEG有损压缩、叠加噪声、剪裁等攻击,具有较好的不可见性和鲁棒性,提高了指纹识别的可靠性。

A fingerprint image watermarking algorithm was proposed based on Contourlet transform and the quantization of parity. After Contourlet transform, original image was decomposed into a series of multi scale, local, and directional subimages. The original binary watermark was firstly scrambled by two-dimensional Arnold transform, and then embedded into the Contourlet transform low frequency sub - band. Then the coefficients of low frequency sub-band were modified by using the quantization of parity. The retrieving watermark algorithm is a blind detecting process, and it does not need original image. The experimental result shows that, this algorithm can resist attacks, such as crop, JPEG compression, and overlaid noise. The watermarking is invisible and robust, and it improves the reliability of fingerprint recognition.