南京大学学报(自然科学版) ›› 2014, Vol. 50 ›› Issue (1): 61–.

• • 上一篇    下一篇

基于时空域能量可疑度的视频篡改检测与篡改区域定位

刘雨青1,黄添强1,2   

  • 出版日期:2014-01-16 发布日期:2014-01-16
  • 作者简介:1. 福建师范大学数学与计算机科学学院,福州,350007; 2. 福建师范大学网络安全与密码技术福建省高校重点实验室,福州,350007
  • 基金资助:
    国家自然科学基金(61070062),福建省高校产学合作科技重大项目(2012H6006),福建省高校服务海西建设重点项目(2008HX200941-4-5),福建省高等学校新世纪优秀人才支持计划(JAI1038),福建省科学厅K类基金项目(JK2011007),福建省教育厅A类基金项目(JA10064)

Digital video forgeries detection and tamper areas location based on temporal and spatial energy suspicious degree

Liu Yuing, Huang Tianiang   

  • Online:2014-01-16 Published:2014-01-16
  • About author:(1. School of Mathematics and Computer ScienceFujian Normal UniversityFuzhou 350007, China 2. Key Laboratory of Network Security and CryptographyFujian Normal University, Fuzhou, 350007, China)

摘要: 针对视频中运动目标从固定背景中移除的情况,提出一种基于能量可疑度计算的视频篡改检测方法,能定位时域和空域上的篡改位置。首先,计算视频各帧能量可疑度,提取时域上的篡改帧序列,再通过帧差法计算可疑运动点图像,提取空域上的可疑运动图像块,根据能量可疑度排除干扰图像块,确定目标移除块,从而实现空域上的篡改定位。实验结果表明,该方法能有效地检测出在固定背景下运动目标是否被移除。

Abstract: In recent years, ith the rapid development of digital multimedia technology, lots of image and video editing software have been widely used which make video become easier to be tampered, so the video tamper detection technology has become a hot research topic in information security field. There are many and varied way of video tampering, and this paper is intended for removing athletic goal from fixed background in the video. A digital video forgery detection method is proposed based on energy suspicious degree calculation, which can track the moving objects by calculating suspicious moving point image, and can locate the temporal and spatial juggled position. Videos are made of “frames”, and the continuity is readily apparent. Therefore, we could transfer video into a series of image in the video tamper detection. The energy suspicious degree is a good representative character which reflects the proportion of the low energy and high energy in each frame. When a moving target is removed from a section of the video image sequence, the grey value, low frequency energy and high frequency energy are bound to change which can affect the energy suspicious degree in the removed area corresponding to the image frame, destroy the continuity and consistency between two adjacent frames, and such change can be used to locate the tamper frame sequences on the time domain. When a goal is deleted, the frame tampering-repair technology mainly uses some patching algorithms to fill the deleted region. The moving objects can be tracked by calculating the suspicious moving point image because of the gray changes in repair area. Measurement procedures are as follow: firstly, extract the juggled frame sequence on the time domain by calculating energy suspicious degree of each frame. Secondly, use the frame differential method to calculate the suspicious moving point image, after that, a statistical method was preferred to extract the spatial suspicious motion image blocks. Finally, set the spatial target removal block as a removing athletic goal and eliminate the spatial interference suspicious blocks by calculating the average of energy suspicious degree in each spatial suspicious block while the spatial tamper location is positioned. The results show that this method can effectively detect whether the moving target was removed from the fixed background and orient the temporal and spatial tamper location accurately.

[1] 陈威兵, 杨高波, 陈日超,等. 数字视频真实性和来源的被动取证. 通信学报, 2011, 32(6):77~182. [2] 周琳娜, 王东明. 数字图像取证技术. 北京: 北京邮电大学出版社, 2008, 8~13. [3] 熊 潇, 黄 征, 徐 彻,等. 基于预测残差检测的数字视频篡改鉴定. 信息安全与通信保密, 2008, 5(12): 128~130. [4] Luo W Q, Wu M, Huang J W. MPEG recompression detection based on block artifacts [C]. Proceedings of the SPIE on Security, Forensics, Steganography, and Watermarking of Multimedia Contents, 2008, 1~12. [5] Wang W H, Farid H. Exposing digital forgeries in video by detecting double MPEG compression [C]. Proceedings of the Multimedia and Security Workshop.Geneva, Switzerland:ACM SIGMM, 2006, 37~47. [6] 黄添强,陈智文.基于双向运动矢量的数字视频篡改鉴定. 山东大学学报(工学版), 2011, 41(4): 13~19. [7] 秦运龙,孙广玲,张新鹏. 利用运动矢量进行视频篡改检测. 计算机研究与发展, 2009, 46(SUPPL): 227~233. [8] 黄添强, 陈智文, 苏立超,等. 利用内容连续性的数字视频篡改检测. 南京大学学报(自然科学), 2011, 47(5): 493~503. [9] 袁秀娟,黄添强,陈智文,等. 基于纹理特征的数字视频篡改检测. 计算机系统应用, 2012, 21(6): 91~95. [10] 张 静, 宋 怡, 苏育挺. 基于空时域联合匹配的视频篡改检测算法. 电子测量技术, 2011, 32(11): 58~69. [11] Zhang J, Su Y T, Zhang M Y. Exposing digital video forgery by ghost shadow artifact [C]. Proceedings of the 1st ACM Workshop on Multimedia in Forensics. Beijing, China, 2009, 49~54. [12] Su Y T, Zhang J, Han Y, et al. Exposing digital video logo-removal forgery by inconsistency of blur [J]. International Journal of Pattern Recognition and Artificial Intelligence, 2010, 24(7): 1027~1046. [13] 陈天华. 数字图像处理. 北京: 清华大学出版社, 2008, 121~123. [14] 赵俊红.篡改图像边缘模糊操作的被动取证算法. 华南理工大学学报(自然科学版), 2011, 39(7): 78~80). [15] 左凤艳,高胜法,韩建宇.基于加权累积差分的运动目标检测与跟踪. 计算机工程, 2009, 35(22): 159~161).
No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!