基于误差扩展与像素容量评估的图像水印算法
An Image Watermarking Algorithm Based on Error Propagation and Pixel Capacity Evaluation

为了解决当前图像水印算法无法准确预测相邻像素间的误差值，不能根据像素的局部纹理特征差异来自适应确定其嵌入容量，导致复原水印信息存在较大失真以及水印容量较小的问题，本文提出了基于误差扩展与像素容量评估的无损图像水印算法.首先，设计动态混淆方法，对水印信息完成加密；再考虑3个相邻像素，引入边缘像素预测机制，准确预测载体图像的每个像素值，获取其对应的误差值；随后，利用载体图像中的任意3个相邻像素的方差，设计像素容量评估方法，对其水印容量完成评估；借助误差扩展技术，设计水印信息嵌入机制，根据像素容量评估结果，将不同容量的水印数据植入到载体图像的不同像素中，获取水印图像；最后，建立水印提取方法，结合解密密钥，准确复原水印信息.实验数据表明：与当前图像水印技术相比，所提算法具有更高的水印质量与水印容量.
In order to solve the defects of the current image watermarking algorithm, such as large distortion of restoration watermark information and small watermark capacity induced by not accurately predicting the error value between neighboring pixels, as well as not adaptively determining its embedding capacity according to the local texture characteristics of pixels, a lossless image watermarking algorithm based on error propagation and pixel capacity evaluation is proposed in this paper. Firstly, a dynamic confusion method is designed to encrypt the watermark information. Next, three adjacent pixels are considered and an edge pixel prediction mechanism is defined to accurately predict each pixel value of the carrier image and to obtain the corresponding error value. Then, a pixel capacity evaluation method is designed according to the variance of any three adjacent pixels in the carrier image to evaluate the watermark embedding capacity of each pixel. The error extension technique is introduced to design the watermark information embedding mechanism for implanting the watermark data with different capacity into different pixels of the carrier image based on the evaluation results of the pixel capacity. Finally, a watermark extraction method is established to recover the watermark information. The results of an experiment show that the proposed algorithm has higher quality of watermarking and watermarking capacity compared with the current image watermarking technology