一种基于人工模糊边缘的图像盲取证方法

Blind image forensics method based on artificially blurred edges

提出一种基于人工模糊边缘检测的盲取证方法,首先基于彩色图像梯度方向特征和边缘邻域斜率特征计算出类似度筛选出疑似模糊边缘点;然后结合图像局部灰度标准差和局部灰度偏导期望值来过滤相机失焦模糊边缘;最后基于线性回归设定阈值过滤掉伪人工模糊边缘。实验结果表明:在减少误判点和区分人工模糊和失焦模糊两方面,本方法在一定程度上具有减少失焦模糊和误判点干扰的优势。

A blind forensics algorithm was proposed based on artificial blurred edges.Firstly,the similarity degree was calculated based on the gradient direction feature of the color image and the slope characteristic of the edge neighborhood,and the suspected blurred edge points were selected.The local grayscale standard deviation of image and the expected value of the local grayscale partial derivative of image were used to filter out-of-focus blurred edges of the camera.Finally,the threshold value was set based on linear regression to filter out false artificial blurred edges.Experimental results show that the algorithm has the advantage of reducing the out-of-focus blur and the interference of mis-judgment points to a certain extent in terms of reducing the misjudgment points and distinguishing between artificial blur and out-of-focus blur.