基于公共因子提取的数字图像篡改检测

摘要

针对数字篡改图像的背景区域和篡改区域都经过模糊的情况,提出了1种基于公共因子提取的模糊篡改检测算法.将图像的线性空间滤波转换成按照行和列方向的一维卷积,通过判断提取出公共因子系数之间的方差来进行篡改检测.该方法特别适用于轻微模糊的情况并且可以同时检测图像的copy_move型篡改.实验说明了该方法对数字图像被动认证的有效性.%A novel forgery image detection scheme for blurred background region and forgery region is proposed in this paper, which determines whether a blurred image is tampered or not using GCD ( greatest common divisor) . Firstly , the image convolution is transformed to one dimensional convolution by row and column directions. Then the tamper is detected by judging the variance between GCD' s coefficients. It particularly suites to minor blur. Simulation results show the efficiency of this approach in revealing the local-blur and copy_move image tamper.