Subventions et des contributions :

Titre :
Deep learning for digital image forensics
Numéro de l’entente :
EGP
Valeur d'entente :
25 000,00 $
Date d'entente :
7 mars 2018 -
Organisation :
Conseil de recherches en sciences naturelles et en génie du Canada
Location :
Colombie-Britannique, Autre, CA
Numéro de référence :
GC-2017-Q4-01867
Type d'entente :
subvention
Type de rapport :
Subventions et des contributions
Informations supplémentaires :

Subvention ou bourse octroyée s'appliquant à plus d'un exercice financier (2017-2018 à 2018-2019).

Nom légal du bénéficiaire :
Wang, Z. Jane (The University of British Columbia)
Programme :
Subventions d'engagement partenarial pour les universités
But du programme :

Digital media data becomes increasingly easier to be recorded, edited and shared. However, this also leads tox000D
serious security and forensics concerns of digital media data. Image forensics has been an active research areax000D
in the past decade, while anti-forensic techniques have also evolved in response to the rapid development ofx000D
image forensics tools because the forgers also attempt to identify weaknesses of image forensics and fool thex000D
forensic investigators by employing anti-forensic techniques. The proposed research project brings togetherx000D
researchers from UBC and Huawei Canada with the goal of developing deep learning based forensic methodsx000D
to learn feature representations from weak forensic trace signals and fulfill classification automatically. Morex000D
specifically, we focus on the following two technical goals:1) develop a deep learning based framework forx000D
image inpainting forensics; and 2) propose a multi-purpose convolutional neural network (CNN) frameworkx000D
for simultaneous detection of multiple types of image manipulations.