通信学报 ›› 2015, Vol. 36 ›› Issue (5): 54-65.doi: 10.11959/j.issn.1000-436x.2015163

• 学术论文 • 上一篇    下一篇

图像镜像复制粘贴篡改检测中的FI-SURF算法

李岩1,刘念2,张斌3,袁开国1,杨义先1   

  1. 1 北京邮电大学 信息安全中心,北京 100876
    2 北京电子科技学院 信息安全系,北京 100070
    3 中国信息安全认证中心,北京 100020
  • 出版日期:2015-05-20 发布日期:2015-07-17
  • 基金资助:
    国家自然科学基金资助项目;新闻出版署基金资助项目;中央高校基本科研业务费专项基金资助项目

FI-SURF algorithm for image copy-flip-move forgery detection

IYan L1,IUNian L2,HANGBin Z3,UANKai-guo Y1,ANGYi-xian Y1   

  1. 1 Information Security Center,Beijing University of Posts and Telecommunications,Beijing 100876,China
    2 Department of Information Security,Beijing Electronic Science and Technology Institute,Beijing 100070,China
    3 China Information Security Certification Center,Beijing 100020,China
  • Online:2015-05-20 Published:2015-07-17
  • Supported by:
    The National Natural Science Foundation of China;General Administration of Press and Publication of the People's Republic of China;Fundamental Research Funds for the Central Universities

摘要:

针对数字图像版权中的复制粘贴篡改问题,提出FI-SURF (flip invariant SURF)算法。研究了当图像经过镜像翻转后SURF (speeded-up robust features)特征描述符的排列变化关系。提取SURF特征点后,将其特征描述符重新排序,即使复制粘贴区域经过镜像翻转,对应的特征点依然可以进行匹配。实验证明,FI-SURF算法在保留SURF算法运算速度快、顽健性强等优点的前提下,可有效检测出经过镜像翻转的复制粘贴区域,计算出复制粘贴区域的轮廓。

关键词: SURF, 特征点, 图像取证, 镜像翻转

Abstract:

FI-SURF (flip invariant SURF) algorithm was proposed,for the consideration of the copy-move forgery detec-tion of digital images.The arrangement of the SURF (speeded-up robust features) descriptor after image flip was studied.After extract the SURF interest points,rearrange the descriptors,the interest points can still be matched even under the condition of copy-flip-move.The experiment results show that FI-SURF not only reserved the advantage of SURF,but also can detect the copy-flip-move forgery,and can locate the duplicated region.

Key words: SURF, interest point, image forgery, copy-flip-move

No Suggested Reading articles found!