Comparative Analysis of a Deep Convolutional Neural Network for Source Camera Identification

Ahmed, Farah, Khelifi, Fouad, Lawgalv, Ashref and Bouridane, Ahmed (2019) Comparative Analysis of a Deep Convolutional Neural Network for Source Camera Identification. In: 2019 IEEE 12th International Conference on Global Security, Safety and Sustainability (ICGS3). IEEE, pp. 1-6. ISBN 978-1-5386-7001-9

Full text not available from this repository. (Request a copy)

Abstract

In image forensics, there is widespread growing investigations for Source Camera Identification (SCI) to identify the legitimate source of an image by using different approaches including filtering-based techniques and deep neural networks. Therefore, the aim of this paper is to shed light on the application of source camera identification in digital forensic science via a comparative analysis of state-of-the art filtered-based technique and deep learning convolutional neural networks-based (CNN). To this end, the basic sensor pattern noise-based estimation technique using the Wiener filter in the wavelet domain is considered as well as deep learning CNN model in order to assess the effectiveness of SCI. Different experiments have been carried out on our database of images taken from eleven different cameras using the same set of training and test images with dimensions 128 × 128 and 256 × 256 for both approaches. From the results of the overall false positive rates, false negative rates and overall accuracy, this paper suggests that filtered-based PRNU approach for SCI outperforms the deep learning CNN approach in digital image forensics when limited number of images are available in forensic investigations.

Item Type:	Book Section
Uncontrolled Keywords:	Filtered-based PRNU, Convolutional Neural Network, Deep Learning, Digital Image Forensics, Source Camera Identification
Subjects:	G400 Computer Science
Department:	Faculties > Engineering and Environment > Computer and Information Sciences
Depositing User:	Becky Skoyles
Date Deposited:	31 May 2019 14:40
Last Modified:	10 Oct 2019 18:45
URI:	http://nrl.northumbria.ac.uk/id/eprint/39434

Downloads