Ecole d'ingénieur et centre de recherche en Sciences du numérique

Image forensics detection by dilated convolutional neural network and error level analysis

Tajini, Badr; Dugelay, Jean-Luc

GdR ISIS/CNRS, 27 September 2018, Paris, France

We live in an era where the digital images are increasingly "photoshopped" before dissemination. Thereby, the quality of the modifications are high even when using basic commercial editing tools that make it easy to create forgeries and that are generally performed for marketing purposes but can also be performed maliciously in the case of fake news spread in social media for example. Thus, It is then important in some cases to be able to detect such forgeries which remains a challenging problem. In this paper, we propose a new method based on dilated convolution neural network that demonstrated very recently high performances in image classification by reducing the misclassification error in ImageNet as well as the location error and the number of network parameters by 94%. In addition, we fed the network with the inconsistent compression information generated by Error Level Analysis (ELA) frequently used with success in image forensics. Our proposed method obtains state-of-the-art performance on four standard image dataset forgery. In addition, the proposed approach includes good robustness against adversarial attacks.

Bibtex

Titre:Image forensics detection by dilated convolutional neural network and error level analysis
Mots Clés:Image Forgery, Image Forensics, Convolutional Neural Network, Dilated Convolution, Error Level Analysis, Deep Learning
Type:Talk
Langue:English
Ville:Paris
Pays:FRANCE
Date:
Département:Sécurité numérique
Eurecom ref:5709
Copyright: © EURECOM. Personal use of this material is permitted. The definitive version of this paper was published in GdR ISIS/CNRS, 27 September 2018, Paris, France and is available at :
Bibtex: @talk{EURECOM+5709, year = {2018}, title = {{I}mage forensics detection by dilated convolutional neural network and error level analysis}, author = {{T}ajini, {B}adr and {D}ugelay, {J}ean-{L}uc}, number = {EURECOM+5709}, month = {09}, institution = {Eurecom} address = {{P}aris, {FRANCE}}, url = {http://www.eurecom.fr/publication/5709} }
Voir aussi: