基于像素不扩展视觉密码的光学彩色脆弱水印

Optical color fragile watermark based on pixel-free expansion visual cryptography

本文提出了一种基于像素不扩展视觉密码的光学彩色脆弱水印系统.一方面,使用像素不扩展视觉密码对水印图像进行编码,避免了因视觉密码引起的像素扩展问题,使得后续可以选择与水印图像具有相同像素大小的彩色宿主图像,大大减少了传输过程中所占用的网络带宽以及存储空间.另一方面,使用相位恢复算法对编码后水印图像进行处理得到用于嵌入宿主图像的相位信息,以光学的方式进一步提高水印图像的安全性.使用计算机模拟验证所提光学彩色脆弱水印的可行性、不可感知性,并通过一系列仿真攻击实验验证所提水印具有良好的脆弱性,在面对噪声污染以及旋转、运动模糊处理、滤波等常见的攻击下均可灵敏地检测出图像发生了篡改.

In recent years,with the continuous development of computer technology,it has brought convenience to people to obtain image information.However,at the same time,the falsification and theft of image information have also emerged,so information security has received increasing attention.When images are used for medicine,military,court,and other purposes,it is necessary to ensure the authenticity and integrity of the image content.Fragile watermarks are used to verify the authenticity and integrity of image content due to their sensitivity to tampering.The watermark information is embedded in the image and integrated with the image.When it is necessary to detect the authenticity and integrity of image information,the extracted watermark can be used to determine whether the image is reliable and complete.Therefore,we propose an optical color fragile watermarking system based on pixel-free expansion visual cryptography.On the one hand,encoding watermark images by using pixel-free expansion visual cryptography avoids pixel expansion issues caused by visual cryptography,allowing for the selection of color host images with the same pixel size as the watermark image in the future,greatly reducing the network bandwidth and storage space occupied during transmission.On the other hand,phase recovery algorithm is used to process the encoded watermark image to obtain phase information for embedding into the host image,further improving the security of the watermark image in an optical way.The feasibility and imperceptibility of the proposed optical color fragile watermark are verified through computer simulation,and its good fragility is verified through a series of simulation attack experiments.It can sensitively detect image tampering in the face of common attacks such as noise pollution,rotation,motion blur processing,filtering,etc.