A Forgery Video Detection Algorithm for Resolution Promotion Manipulations Using Frequency Spectrum Analysis

International Conference on Network, Communication and Computing Pub Date : 2016-12-17 DOI:10.1145/3033288.3033304

S. Song, Heng Yao, Chuan Qin, Ying Tian

{"title":"A Forgery Video Detection Algorithm for Resolution Promotion Manipulations Using Frequency Spectrum Analysis","authors":"S. Song, Heng Yao, Chuan Qin, Ying Tian","doi":"10.1145/3033288.3033304","DOIUrl":null,"url":null,"abstract":"Resolution is one of the important factors for video visual quality. With the development of the video editing software, one video sequence can be easily enlarged and inserted into another video sequence to generate a forgery video. The detection of video resolution modifications has become an important issue for digital video authentication. In this paper, we propose a passive detection algorithm based on the zero-crossings of the high-frequency component of the key frames in frequency domain. Firstly, the key frames of video are extracted before the second-order differential is executed. Then, the zero crossings of differential are calculated to generate a one-dimensional binary sequence. Finally, the frequency analysis is applied to expose the original resolution of the video for detection. Based on this method, we also develop a forgery detection method for sequence inserting. The experimental results demonstrate that the proposed method can accurately estimate the video original resolution, as well as detect the forgery traces of synthetic videos with maliciously inserted irrelevant video sequences.","PeriodicalId":253625,"journal":{"name":"International Conference on Network, Communication and Computing","volume":"14 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2016-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Conference on Network, Communication and Computing","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/3033288.3033304","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 0

Abstract

Resolution is one of the important factors for video visual quality. With the development of the video editing software, one video sequence can be easily enlarged and inserted into another video sequence to generate a forgery video. The detection of video resolution modifications has become an important issue for digital video authentication. In this paper, we propose a passive detection algorithm based on the zero-crossings of the high-frequency component of the key frames in frequency domain. Firstly, the key frames of video are extracted before the second-order differential is executed. Then, the zero crossings of differential are calculated to generate a one-dimensional binary sequence. Finally, the frequency analysis is applied to expose the original resolution of the video for detection. Based on this method, we also develop a forgery detection method for sequence inserting. The experimental results demonstrate that the proposed method can accurately estimate the video original resolution, as well as detect the forgery traces of synthetic videos with maliciously inserted irrelevant video sequences.

查看原文本刊更多论文

一种基于频谱分析的分辨率提升伪视频检测算法

分辨率是影响视频视觉质量的重要因素之一。随着视频编辑软件的发展，一个视频序列可以很容易地放大并插入到另一个视频序列中，从而产生伪造视频。视频分辨率修改检测已成为数字视频认证中的一个重要问题。本文提出了一种基于关键帧高频分量的频域过零的被动检测算法。首先，在进行二阶微分之前提取视频的关键帧;然后，计算微分函数的过零点，得到一维二进制序列。最后，利用频率分析来暴露视频的原始分辨率以供检测。在此基础上，我们还开发了一种序列插入伪造检测方法。实验结果表明，该方法能准确估计视频的原始分辨率，并能检测出恶意插入不相关视频序列的合成视频的伪造痕迹。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

International Conference on Network, Communication and Computing

自引率

0.00%

发文量