文献类型：期刊文献

英文题名：Image copy-move forgery detecting based on local invariant feature

作者：Jing, Li[1]; Shao, Chao[1]

第一作者：景丽

通讯作者：Jing, L.|[1048412295f21ea]景丽;

机构：[1] School of Computer and Information engineering, Henan University of Economics and Law, Zhengzhou, China

第一机构：河南财经政法大学计算机与信息工程学院

通讯机构：[1]School of Computer and Information engineering, Henan University of Economics and Law, Zhengzhou, China|[1048412]河南财经政法大学计算机与信息工程学院;[10484]河南财经政法大学;

年份：2012

卷号：7

期号：1

起止页码：90-97

外文期刊名：Journal of Multimedia

收录：EI(收录号：20121014833825);Scopus(收录号：2-s2.0-84857593016)

语种：英文

外文关键词：Digital storage - Forestry - Image compression - Motion compensation - Trees (mathematics)

摘要：Now digital images are widely used in many fields. Making image forgeries with digital media editing tools is very easy, and these image forgeries are undetectable by human eyes. Copy-move forgery is common image tampering where a part of the image is copied and pasted on another parts. Up to now the useful way to detect copy-move forgeries is block matching technique. This paper firstly analyzes and summarizes block matching technique, then introduces a copy-move forgery detecting method based on local invariant feature matching. It locates copied and pasted regions by matching feature points. It detects feature points and extracts local feature using Scale Invariant Transform algorithm. Matching local features is based on k-d tree and Best-Bin-First method. Through analysis we learn computational complexity of the proposed method is similar to existing block-matching methods, but has better locating accuracy. Experiments show that this method can detect copied and pasted regions successively, even when these regions are operated by some process, such as JPEG compression, Gaussian blurring, rotation and scale. ? 2012 Academy Publisher.