摘要
区域复制把数字图像中一部分区域进行复制并粘贴到同一幅图像的另一个区域中,以达到去除图像中某一重要内容的目的,是一种简单而有效的图像篡改技术.现有检测算法对区域复制后处理的鲁棒性较差.文中针对此篡改技术,提出了一种有效的检测与定位篡改区域算法.该算法首先将图像分解为小块并比较各小块间的相似性,最后利用"主转移向量"方法去除错误的相似块对得到篡改的区域.实验数据说明该算法能有效地对抗多种区域复制的后处理操作,包括高斯模糊、加性白高斯噪声、JPEG压缩及它们的混合操作.
Region duplication forgery, in which a part of a digital image is copied and then pasted to another portion of the same image in order to conceal an important object in the scene, is one of the common image forgery techniques. This paper describes an efficient and robust algorithm for detecting and localizing this type of malicious tampering. The algorithm first divides an image into small overlapped blocks and then compares the similarity of these blocks, finally identifies possible duplicated regions using the main shift vector. The experimental results show that above method is more robust comparing with other existing algorithms and can successfully detect this type of tampering for images that have been subjected to various forms of post region duplication image processing, including, blurring, noise contamination, severe lossy compression, and a mixture of these processing operations.
出处
《计算机学报》
EI
CSCD
北大核心
2007年第11期1998-2007,共10页
Chinese Journal of Computers
基金
国家杰出青年科学基金(60325208)
国家自然科学基金(60633030
90604008)
国家"九七三"重点基础研究发展规划项目基金(2006CB303104)
广东省自然科学基金团队项目(04205407)资助.~~
关键词
图像盲认证
图像篡改
篡改检测
区域复制
鲁棒性
blind image authentication
image forgery
detection of tampering
region duplication
robustness
