视频帧内运动目标复制-粘贴篡改检测算法

上海交通大学学报（自然版） ›› 2015, Vol. 49 ›› Issue (02): 196-201.

• 自动化技术、计算机技术 • 上一篇下一篇

视频帧内运动目标复制-粘贴篡改检测算法

张璐波a，孙锬锋a,b,c，蒋兴浩a,b,c

(上海交通大学a.电子信息与电气工程学院； b. 信息内容分析技术国家工程实验室；c. 上海市信息安全综合管理技术研究重点实验室, 上海 200240)

收稿日期:2014-06-11 出版日期:2015-02-28 发布日期:2015-02-28
基金资助:
国家自然科学基金（61272249，61272439），高等教育博士点专项基金(20120073110053)，上海市科委国际合作项目(12510708500)资助

Moving Targets CopyMove Forgery Detection in Video Sequences

ZHANG Luboa,SUN Tanfenga,b,c,JIANG Xinghaoa,b,c

(a. School of Electronic Information and Electrical Engineering; b. National Engineering Laboratory on Information Content Analysis Techniques; c. Shanghai Key Laboratory of Administration Technologies for Information Security, Shanghai Jiaotong University, Shanghai 200240, China)

Received:2014-06-11 Online:2015-02-28 Published:2015-02-28

摘要/Abstract

摘要：

摘要：提出了一种基于Lucas_Kanade(LK)光流及运动目标预检机制的视频帧内运动目标复制粘贴篡改检测算法.该算法分为运动目标检测与跟踪、运动序列筛选和空间域匹配3个阶段.运动目标检测与跟踪利用背景建模算法和卡尔曼滤波器进行检测和跟踪；运动序列的筛选采用LK方法得到各运动序列的光流值，并计算其相关性来选择可能存在篡改的视频帧序列；空间域匹配利用尺度不变特征变换算法对上一阶段得到的对应运动序列逐帧进行匹配，过滤正常的视频序列.实验结果表明，本文算法能有效检测同源视频中针对运动目标的多帧复制粘贴篡改.

关键词: 视频篡改检测, 运动目标检测跟踪, 光流算法, 尺度不变特征, 数字视频取证

Abstract:

Abstract： In this paper, a novel technique to detect moving targets copy-move forgery in video sequences based on LK optical flow and the pre-screening mechanism of moving targets was proposed. The proposed scheme is a coarse-to-fine approach and composed of moving targets detection and tracking, candidate motion sequences selection, and spatial matching. In the pre-processing round, the visual background extractor (ViBe) mechanism and Kalman filter were used to detect and tracking moving objects. To screen duplicated candidates in the temporal domain, the optical flow features in each motion sequences were obtained using the Lucas_Kanade algorithm and calculate their correlation. To evaluate the similarity of image content, the scale invariant feature transform (SIFT) algorithm was used to measure spatial correlation of each corresponding frame between the motion sequences. The experimental results show that the proposed scheme can detect moving targets copy-move forgery well in homologous video.
Key words：

Key words: video tampering detection, moving objects detection and tracking, optical flow algorithm, scale invariant feature, digital video forensics

中图分类号:

TP 391

张璐波a，孙锬锋a,b,c，蒋兴浩a,b,c. 视频帧内运动目标复制-粘贴篡改检测算法[J]. 上海交通大学学报（自然版）, 2015, 49(02): 196-201.

ZHANG Luboa,SUN Tanfenga,b,c,JIANG Xinghaoa,b,c. Moving Targets CopyMove Forgery Detection in Video Sequences[J]. Journal of Shanghai Jiaotong University, 2015, 49(02): 196-201.

参考文献

［1］Milani S, Fontani M, Bestagini P, et al. An overview on video forensics［J］. APSIPA Transactions on Signal and Information Processing, 2012, 1(1): 118.

［2］Wang W, Farid H. Exposing digital forgeries in video by detecting double MPEG compression ［C］∥Proceedings of the 8th Workshop on Multimedia and Security. New York, USA:ACM, 2006:3747.

［3］Wang W, Farid H. Exposing digital forgeries in video by detecting double quantization ［C］∥Proceedings of the 11th ACM Workshop on Multimedia and Security. New York, USA: ACM, 2009:3948.

［4］Wang W, FARID H. Exposing digital forgeries in video by detecting duplication ［C］∥International Multimedia Conference on Proceeding of the 9th Workshop on Multimedia and Security. Dallas, Texas, USA:ACM, 2007: 3542.

［5］Chao J, Jiang X H, Sun T F. A new video interframe tampering detection method based on optical flow consistency ［C］∥The 11th International Workshop on Digitalforensics and Watermarking. Shanghai: Springer, 2012: 2527.

［6］Stamm M C, Lin W S K, Liu J R. Temporal forensics and antiforensics for motion compensated video ［J］. IEEE Transaction on Information Forensics and Security, 2012, 7(4): 13151329.

［7］李富贵，黄添强，苏立超, 等.融合多特征的异源视频复制粘贴篡改检测［J］.山东大学学报(工学版），2013，43(4): 3238.

LI Fugui, HUANG Tianqiang, SU Lichao, et al. Heterologous video copymove forgery detection by fusing multiple features［J］. Journal of Shandong University (Engineering Science), 2013, 43(4): 3238.

［8］Bestagini P, Milani S, Tagliasacchi M, et al, Local tampering detection in video sequences［C］∥2013 IEEE 15th International Workshop on Multimedia Signal Processing (MMSP). Pula, Italy:IEEE, 2013:488493.

［9］Subramanyam A V, Emmanuel S. Video forgery detection using HOG features and compression properties［C］∥14th IEEE International Workshop on Multimedia Signal Processing (MMSP). Banff, Canada:IEEE, 2012:8994.

［10］Barnich O, Van Droogenbroeck M. ViBe: A universal background subtraction algorithm for video sequences ［J］. IEEE Transactions on Image Processing, 2011, 20(6):17091724.

［11］Welch G, Bishop G.An introduction to the Kalman filter［C］∥SIGGRAPH 2001. Los Angeles, CA: ACM, 2001: 1217.

［12］Lucas B D, Kanade T. An iterative image registration technique with anapplication to stereo vision［C］∥IJCAI. San Francisco, CA USA: Publisher Morgan Kaufmann Publishers Inc, 1981:674679.

［13］Qadir G,Yahaya S, Ho A T S. Surrey university library forforensic analysis (SULFA) of video content ［C］∥IET Conference on Image Processing (IPR 2012). London:IET, 2012: 16.

[1]	刘士荣1,2，吴楚1,3，张波涛1,2，章海兵1,2. 基于极线约束SIFT特征和粒子滤波的目标跟踪算法[J]. 上海交通大学学报（自然版）, 2014, 48(07): 1026-1032.
[2]	巢娟1, 孙锬锋1, 2, 蒋兴浩1, 2. 基于双重检测模型的视频镜头分割算法[J]. 上海交通大学学报（自然版）, 2011, 45(10): 1542-1546.

视频帧内运动目标复制-粘贴篡改检测算法

Moving Targets CopyMove Forgery Detection in Video Sequences

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 2

编辑推荐

Metrics

本文评价