A Two-stage Algorithm for Video Forgery Detection Based on Inception-V3 Network

Weng Shao-wei; Peng Yi-hang; Wei Bo; Yi Lin; Ye Wu-jian

doi:10.12052/gdutxb.190046

English Version Journal Login

您当前的位置：

首页 >

文章列表页 >

A Two-stage Algorithm for Video Forgery Detection Based on Inception-V3 Network

更新时间：2026-05-27

- A Two-stage Algorithm for Video Forgery Detection Based on Inception-V3 Network
- Vol. 36, Issue 6, Pages: 16-23(2019)
- 作者机构：
  
  1. 广东工业大学　信息工程学院,广东,广州,510006
  2. 广东省智能信息重点实验室,广东,深圳,518060
- 作者简介：
- 基金信息：
- DOI：10.12052/gdutxb.190046
  CLC： TP391
- Published：2019
- 稿件说明：
移动端阅览
翁韶伟, 彭一航, 危博, 易林, 叶武剑. 基于Inception-V3网络的双阶段数字视频篡改检测算法[J]. 广东工业大学学报, 2019, 36(6): 16-23.

Weng Shao-wei, Peng Yi-hang, Wei Bo, et al. A Two-stage Algorithm for Video Forgery Detection Based on Inception-V3 Network[J]. 2019, 36(6): 16-23.
翁韶伟, 彭一航, 危博, 易林, 叶武剑. 基于Inception-V3网络的双阶段数字视频篡改检测算法[J]. 广东工业大学学报, 2019, 36(6): 16-23. DOI： 10.12052/gdutxb.190046.

Weng Shao-wei, Peng Yi-hang, Wei Bo, et al. A Two-stage Algorithm for Video Forgery Detection Based on Inception-V3 Network[J]. 2019, 36(6): 16-23. DOI： 10.12052/gdutxb.190046.

摘要

为了克服现有数字视频取证算法识别准确率低、定位能力差等缺点

提出一种具有高识别率且定位准确的基于Inception-V3网络的二级分类取证算法. 在第一级分类器中提出简单的阈值判断方法来区分原始和篡改视频

第二级分类器将采用Inception-V3网络的稠密卷积核结构来自动提取篡改视频帧的高维多尺度特征. 高维多尺度特征有助于提升篡改视频帧的识别率. 实验结果表明

本文提出的算法不仅能准确地检测出篡改视频

还能从篡改视频中精确定位出篡改帧.

Abstract

In order to overcome the shortcomings (e.g. low recognition accuracy and poor localization ability) of existing video forensics techniques

a two-stage algorithm is proposed based on Inception-V3 network

whose advantage lies in the fact that it can accurately identify a forged video and locate its forged frames. After an extensive research

it is found that the average value of all the pixels in a pristine video sequence is always larger than its forged one after the video sequence is processed via several operations such as high-pass filtering and convolution. To this end

in the first stage

a simple algorithm is proposed in which a predefined threshold is employed to distinguish a forged video and a pristine video. Considering the fact that features of each frame need to be extracted manually in existing algorithms

the dense convolution kernel structure of Inception-V3 network is adopted in the second stage to automatically extract high dimensional and multi-scale features of each forged frame. Inception-V3 network can accurately locate forged frames in a forged video since dimensional and multi-scale features can more adequately express the input information. The experiments show that the proposed method performs very well both in forged video identification and forged frame localization.

关键词

Keywords

references

NA S, OH W, JEONG D. A frame-based video signature method for very quick video identification and location [J]. ETRI Journal, 2013, 35(2): 281-291.

LEE S, YOO C D. Robust video fingerprinting for content-based video identification [J]. IEEE Trans on Circuits and Systems for Video Technology, 2008, 18(7): 983-988.

CHETTY G, BISWAS M, SIGNH R. Digital video tamper detection based on multimodal fusion of residue features[C] //Fourth International Conference on Network System Security. IEEE Computer Society. Melbourne: IEEE, 2010: 606-613.

GOODWIN J, CHETTY G. Blind video tamper detection based on fusion of source features[C] //International Conference on Digital Image Computing: Techniques Applications. Noosa: IEEE, 2011: 608-613.

BESTAGINI P, MILANI S, TAGLIASACCHI M, et al. Local tampering detection in video sequences[C] //Proceedings of the IEEE 15th International Workshop on Multimedia Signal Processing. Pula: IEEE, 2013: 488–493.

LI L, WANG X, ZHANG W, et al. Detecting removed object from video with stationary background[C] //International Conference on Digital Forensics Watermarking. Shanghai: Springer, 2012: 242-252.

BIDOKHTI A, CHAEMMAGHAMI S. Detection of region copy move forgery in MPEG videos using optical flow[C] // International Symposium on Artificial Intelligence and Signal Processing. Mashhad: IEEE, 2015: 13-17.

PANDEY R C, SINGH S K, SHUKLA K K. Passive copy-move forgery detection in videos[C] //Proceedings of the International Conference on Computer and Communication Technology. Allahabad: Springer, 2014: 301–306.

YE W J, GAO H J, WENG S W, et al . A Two-sta ge effect rendering method for art font based on CGAN network [J ] . Journal of Guangdong University of Technology, 2019, 36(3): 48-54.

CHEN S, TAN S, LI B, et al . Automatic detection of object-based forgery in advanced video [J ] . IEEE Trans on Circuits and Systems for Video Technology, 2016, 26(11): 2318-2350.

YAO Y, SHI Y, WENG S, et al . Deep learning for detection of object-based forgery in advanced video [J ] . Symmetry, 2017, 10(3): 1-10.

LIU S H, HAN L, YAO H X. Video watermarking algorithm for resisting collusion attacks [J]. Journal on Communications, 2010, 31(1): 14-19.

CHRISTIAN S, VICENT V, LOFFE S, et al. Rethinking the inception architecture for computer vision[C] //IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas: IEEE, 2016: 2818-2826.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Efficient Temporal Modeling for RGB-T Tracking

Energy Management Strategy of Hybrid Locomotive Based on Road Condition Parameters

Modeling of 3D NAND Memory Cells with Taper Angle and Programming Strategy Optimization

A Resource Allocation Optimization Method for Multi-UAV Integrated Sensing, Computing, and Communication Towards Disaster Relief

Research and Implementation of a Pulsed Laser Ranging Method Based on Digital Mixing

Related Author

No data

Related Institution

No data

AI问答

⁰