结合边缘增强的零水印

Zero-watermarking algorithm using edge enhancing

鲁棒性和不可见性是水印的两个重要特性。为实现该目的,利用奇异值的稳定性和图像梯度提出了一种新的零水印算法。该算法先对图像做分块奇异值分解,提取每块的最大奇异值构成原图像的缩略图;然后计算缩略图的梯度模值并二值化,得到边缘增强的二值图像,即图像的特征信息。最后将处理过的水印信息与图像特征信息做异或运算,其结果将作为提取水印的密钥。实验证明,本算法对滤波、锐化、直方图均衡、压缩、剪切等图像处理操作具有很强的鲁棒性。

Robustness and invisibility are two important characteristics of watermarking. To meet these requirements, a new zero-watermarking algorithm based on Singular Value Decompositon （SVD） and gradient was proposed in this paper. Firstly, do block-SVD transform on the host image. Then select the largest singular value of each block. After that a small scale image which is similar to the host image will be obtained. Secondly calculate the small scale image＇ s gradient modulus, and divide these data into two parts using the following rules： if the value is greater than the mean, the result is 1 ; or is 0. The consequence is the image＇ s characteristic information. Finally do XOR operation on the characteristic data and the watermark data. When detecting the watermark, the result of XOR operation will be used as the key. The experimental results show that this approach has good performance to the filtering, sharpening, histogram equalization, JPEG compression and crop.