A Convolutional Neural Network Ensemble for Video Source Camera Forensics
IEEE MultiMedia · 2024-04 · 31 (2) · 26-35
DOI: 10.1109/mmul.2024.3372372 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE MultiMedia, 2024-04. Verfasst von Maryna Veksler und 3 weiteren Autor:innen. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 20 Literaturangaben. Gefördert von: U.S. Army Research Office.
Abstract (Originalsprache)
The advancement of Internet of Things technologies has influenced a substantial increase in the use of multimedia devices. Consequently, forensic specialists have begun to experience a vast load of video and image data during investigative procedures. This has triggered a need to ensure the integrity of multimedia data and verify its source origin for digital forensics processes. In this article, we address the problem of identifying the video source camera of the video data acquired by investigators. We develop a novel convolutional neural network (CNN) ensemble framework to identify the video source camera. In our method, we analyze the video data using patches extracted from intracoded frame (I-frame) quadrants (i.e., nonoverlapping squares) using independent CNNs for each quadrant to achieve location awareness. Experimental results demonstrate that our framework is robust for the same device-type classification and outperforms existing deep learning-based techniques.
Förderung
- U.S. Army Research Office
Zitieren
APA
Veksler, M., Aygun, R. S., Akkaya, K., & Iyengar, S. S. (2024). A Convolutional Neural Network Ensemble for Video Source Camera Forensics. IEEE MultiMedia, 31(2), 26-35. https://doi.org/10.1109/mmul.2024.3372372
Vancouver
Veksler M, Aygun RS, Akkaya K, Iyengar SS. A Convolutional Neural Network Ensemble for Video Source Camera Forensics. IEEE MultiMedia. 2024;31(2):26-35. doi:10.1109/mmul.2024.3372372
BibTeX
@article{Veksler2024,
author = {Veksler, Maryna and Aygun, Ramazan S. and Akkaya, Kemal and Iyengar, Sitharama S.},
title = {A Convolutional Neural Network Ensemble for Video Source Camera Forensics},
journal = {IEEE MultiMedia},
year = {2024},
volume = {31},
number = {2},
pages = {26--35},
doi = {10.1109/mmul.2024.3372372}
}
EUVIMED