Title: Multimodal Fingerprinting of Imaging Devices – Inception Report
Authors: FERRARA PASQUALEBESLAY Laurent
Publisher: Publications Office of the European Union
Publication Year: 2016
JRC N°: JRC105300
ISBN: 978-92-79-64956-1
ISSN: 1831-9424
Other Identifiers: EUR 28402 EN
OP KJ-NA-28402-EN-N
URI: http://publications.jrc.ec.europa.eu/repository/handle/JRC105300
DOI: 10.2760/751517
Type: EUR - Scientific and Technical Research Reports
Abstract: This report aims at surveying the state-of-art regarding source camera identification techniques, that can be conceived as complementary to the Sensor Pattern Noise (SPN) based method, that has been already developed within the framework of AVICAO (Authors and Victims Identification of Child-Abuse Online) JRC project in collaboration with EUROPOL European Cyber Crime Centre (EC3). The purpose of AVICAO project is to boost European LEAs capabilities in fighting against Child-Abuse Online. In such context, the topic of multimodal source camera identification is a new research field of the activities carried out within the named project and consists of the finding, evaluating, selecting and integrating of new features able to come up beside the SPN in identifying source cameras of still images and video recordings. Furthermore, JRC has identified such an application suitable for combatting another serious crime, namely terrorism, whose videos are shared on the World Wide Web for propaganda purposes. The report reviews the most promising techniques available in literature, analysing in depth their properties in term of uniqueness in identifying devices, accuracy performance and stability over time. Furthermore, decision fusion techniques are also explored to provide an effective practice to combine decisions from different classifiers.
JRC Directorate:Space, Security and Migration

Files in This Item:
File Description SizeFormat 
jrc105300_09_01_ipo.pdf791.91 kBAdobe PDFView/Open


Items in repository are protected by copyright, with all rights reserved, unless otherwise indicated.