灰度差异耦合显著特征的图像复制-粘贴篡改检测算法  

An image copy-paste tampering detection algorithm based on gray difference coupled salient feature

作　　者：刘震 刘容志[2] LIU Zhen;Liu Rongzhi(School of Information Engineering,Wuhan College,Wuhan 430212,China;School of Business Administration,Zhongnan University of economics and law,Wuhan 430212,China)

机构地区：[1]武汉学院信息工程学院,湖北武汉430212 [2]中南财经政法大学工商管理学院,湖北武汉430212

出　　处：《光学技术》2020年第5期619-625,共7页Optical Technique

基　　金：国家自然科学基金面上项目(71373251);国家社会科学基金项目青年项目(16CGL005);湖北产业政策与管理研究中心项目(CY20150209)。

摘　　要：当前较多图像篡改检测算法主要是依靠测量图像间的距离来匹配像素特征,以此来实现图像的篡改检测,但其没有考虑图像特征间显著特征的关联性,导致检测结果存在误检等弊端。为此,设计了灰度差异耦合显著特征的图像复制-粘贴篡改检测算法。首先,借助Harris算子,从待检测图像中提取初始特征,并利用像素点间的灰度差异信息,删除伪特征,以得到准确度较高的图像特征。然后,将Haar小波作为求取图像特征方向信息的依据,利用图像锐度信息,获取特征向量。采用傅里叶变换来计算图像特征的显著信息,利用其对图像特征实施匹配。最后,对匹配图像特征进行聚类计算,以辨别出图像中的篡改内容。实验数据表明,较当前伪造检测算法而言,所提算法不仅具有更少的错误检测内容,而且拥有更强的健壮性,能抵御多种仿射变换对篡改检测的干扰。At present,many image tamper detection algorithms mainly rely on measuring the distance between images to match the pixel features for achieving image tamper detection,but it does not consider the correlation between the significant features of the image features,which resulting in the defects as false detection in the detection results.For this reason,an image copy paste tamper detection algorithm based on gray difference and salient features is proposed.Firstly,the initial features were extracted from the image to be detected by using the haaris operator,and the pseudo features were deleted by using the gray difference information between pixels to get the image features with high accuracy.Then,the Haar wavelet was used as the basis to get the direction information of image feature,and the feature vector was obtained by using the sharpness information of image.The Fourier transform was used to calculate the salient information of image features for matching image features.Finally,the matching image features are clustered to identify the tampered content in the image.Experimental data show that compared with the current forgery detection algorithm,the proposed algorithm not only has less error detection content,but also has stronger robustness,which can resist the interference of a variety of affine transformations on tamper detection.

关 键 词：图像篡改检测 HARRIS算子 图像特征 锐度信息 显著特征 灰度差异信息 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]