Abstract
Video forensics becomes more and more important than ever before. In this paper a new methodology based on Block-wise Brightness Variance Descriptor (BBVD) is proposed. It is capable of fast detecting video inter-frame forgery. Our proposed algorithm has been tested on a database consisting of 240 original and forged videos. The experiments have demonstrated that the precision rate is about 94.09 % in detecting the insertion forgery and the precision rate is 79.45 % in the forgery localization. Moreover, the time utilized for forgery detecting is shorter than the time used for video replay. On average the time of forgery detection is only about 73.4 % in video replay.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 18-30 |
| Number of pages | 13 |
| Journal | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
| Volume | 9023 |
| DOIs | |
| State | Published - 2015 |
| Event | 13th International Workshop on Digital-Forensics and Watermarking , IWDW 2014 - Taipei, Taiwan, Province of China Duration: Oct 1 2014 → Oct 4 2014 |
All Science Journal Classification (ASJC) codes
- Theoretical Computer Science
- General Computer Science
Keywords
- Block-wise brightness variance descriptor
- Inter-frame forgery
- Video forensics