| 一种基于特征提取与追踪的视频复制粘贴篡改检测方法 |
| |
| 引用本文： | 苏立超,黄添强,郑明魁.一种基于特征提取与追踪的视频复制粘贴篡改检测方法[J].福州大学学报(自然科学版),2015,43(4):450-455. |
| |
| 作者姓名： | 苏立超 黄添强 郑明魁 |
| |
| 作者单位： | 福建师范大学数学与计算机科学学院,福建师范大学数学与计算机科学学院,福州大学物理与信息工程学院 |
| |
| 基金项目： | 国家自然科学基金(61070062)、福建省高校服务海西建设重点项目(No.2008HX200941-4-5)、福建省高校新世纪优秀人才项目(JA11038)、福建省高校创新团队培育计划 |
| |
| 摘 要： | 提出一种基于特征提取与追踪的视频篡改检测方法.首先对视频序列按照一定帧间隔进行特征提取,寻找出首个篡改帧及其篡改区域;其次采用追踪算法对该篡改区域进行学习,并定位该帧之后的每个视频帧的可疑区域;最后对可疑区域再使用特征提取算法进行相似点匹配,确定出视频篡改帧的位置及其篡改区域.结果证明了方法的有效性和鲁棒性,以及在执行时间上相对于已有算法的优越性.
|
| 关 键 词： | 视频篡改 特征提取 压缩追踪 视频帧 |
A copy-paste video tamper detection method based on features extracting and tracking |
| |
| SU Lichao,HUANG Tianqiang and ZHENG Mingkui.A copy-paste video tamper detection method based on features extracting and tracking[J].Journal of Fuzhou University(Natural Science Edition),2015,43(4):450-455. |
| |
| Authors: | SU Lichao HUANG Tianqiang and ZHENG Mingkui |
| |
| Institution: | School of Mathematics and Computer Science of Fujian Normal University,Fuzhou,China College of Physics and Information of Fuzhou University,Fuzhou,China,School of Mathematics and Computer Science of Fujian Normal University,Fuzhou,China College of Physics and Information of Fuzhou University,Fuzhou,China,College of Physics and Information Engineering,Fuzhou University,Fuzhou,China |
| |
| Abstract: | Intra-frame copy-paste of video is one of the common methods of video tampering. In this paper, a novel video tamper detection method based on features extracting and tracking is proposed. In the method, the SIFT algorithm is firstly used to extract features of video sequence according to inspection interval parameter and seek out the first tampering frame and the tampered area. Then, the compression tracking algorithm is used to study the tampered area and locate the suspicious area of each video frame. Finally, the SIFT algorithm is used to calculate the similarity of suspicious area obtained and the position of tampered frames of the whole video and their specific tampered areas are confirmed. The experimental result show that this method are more effective, robust and quick than that of before. |
| |
| Keywords: | Video forgery Feature extracting Compressive tracking Video frame |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《福州大学学报(自然科学版)》浏览原始摘要信息 |
| 点击此处可从《福州大学学报(自然科学版)》下载免费的PDF全文 |
|
