Author Login Editor-in-Chief Peer Review Editor Work Office Work

Computer Engineering

Previous Articles    

Detection of Video Inter Frame Forgery Based on Texture Spectrum

LIN Xinqi 1,2,LIN Yunmei 1,LIN Zhixin 3,KONG Xiangzeng 1,2,YAN Xiaoming 1,2   

  1. (1.School of Mathematics and Computer Science,Fujian Normal University,Fuzhou 350007,China; 2.Fujian Provincial Key Laboratory of Network Security and Cryptology,Fuzhou 350117,China; 3.School of Information and Software Engineering,University of Electronic Science and Technology of China,Chengdu 611731,China)
  • Received:2014-09-15 Online:2015-11-15 Published:2015-11-13

基于纹理谱的视频帧间篡改检测

林新棋 1,2,林云玫 1,林志新 3,孔祥增 1,2,严晓明 1,2   

  1. (1.福建师范大学数学与计算机科学学院,福州 350007; 2.福建省网络安全与密码技术重点实验室,福州 350117; 3.电子科技大学信息与软件工程学院,成都 611731)
  • 作者简介:林新棋(1972-),男,副教授、博士,主研方向:多媒体技术,编码理论;林云玫、林志新,硕士研究生;孔祥增,实验师、博士研究生;严晓明,副教授、硕士。
  • 基金资助:
    福建省教育厅基金资助项目(JA12075,JA10064,JB11036);福建省科技厅高校产学合作科技基金资助重大项目(2012H6006);福建省高等学校科技创新团队基金资助项目(IRTSTFJ,J1917);福建师范大学创新团队基金资助项目(IRTL1207)。

Abstract: In order to improve the accuracy and time efficiency of the tamper video detecting algorithm,a new algorithm based on texture spectrum is proposed.The algorithm is implemented in three stages:firstly,the spectrum image sequences of the video clip is computed;then,the correlation coefficient and abnormality degree of texture spectrum image sequences are defined and calculated;finally,the forgery places are detected by threshold approach.Experimental results show that the proposed method can effectively localize the tamped position of the frame duplication,insertion,deletion tampers.The average consumption of time is 0.4 s per frame in resolution 352×288.The precision and recall are all exceeding 94%.

Key words: texture spectrum image, video forgery, correlation coefficient, abnormality degree, threshold

摘要: 为提高视频篡改检测算法的准确率,提出一种基于纹理谱的时间域视频篡改检测算法。计算视频片段的纹理谱图像序列的相关系数,根据阈值法判定视频篡改位置。实验结果表明,该算法能准确检测出视频帧删除、复制以及插入3种篡改方式。在分辨率为352×288的 视频片段上,平均每帧处理时间不超过0.4 s,而查准度和查全率全部超过94%。

关键词: 纹理谱图像, 视频篡改, 相关系数, 异常度, 阈值

CLC Number: