Please use this identifier to cite or link to this item: https://scholarhub.balamand.edu.lb/handle/uob/6839
DC FieldValueLanguage
dc.contributor.authorTekli, Jimmyen_US
dc.contributor.authorAl Bouna, Becharaen_US
dc.contributor.authorTekli, Gilberten_US
dc.contributor.authorCouturier, Raphaëlen_US
dc.date.accessioned2023-06-06T06:18:34Z-
dc.date.available2023-06-06T06:18:34Z-
dc.date.issued2023-04-11-
dc.identifier.issn13807501-
dc.identifier.urihttps://scholarhub.balamand.edu.lb/handle/uob/6839-
dc.description.abstractImage obfuscation techniques (e.g., pixelation, blurring and masking,..) have been developed to protect sensitive information in images (e.g. individuals’ faces). In a previous work, we designed a recommendation framework that evaluates the robustness of image obfuscation techniques and recommends the most resilient obfuscation against Deep-Learning assisted attacks. In this paper, we extend the framework due to two main reasons. First, to the best of our knowledge there is not a standardized evaluation methodology nor a defined model for adversaries when evaluating the robustness of image obfuscation and more specifically face obfuscation techniques. Therefore, we adapt a three-components adversary model (goal, knowledge and capabilities) to our application domain (i.e., facial features obfuscations) and embed it in our framework. Second, considering several attacking scenarios is vital when evaluating the robustness of image obfuscation techniques. Hence, we define three threat levels and explore new aspects of an adversary and its capabilities by extending the background knowledge to include the obfuscation technique along with its hyper-parameters and the identities of the target individuals. We conduct three sets of experiments on a publicly available celebrity faces dataset. Throughout the first experiment, we implement and evaluate the recommendation framework by considering four adversaries attacking obfuscation techniques (e.g. pixelating, Gaussian/motion blur and masking) via restoration-based attacks. Throughout the second and third experiments, we demonstrate how the adversary’s attacking capabilities (recognition-based and Restoration & Recognition-based attacks) scale with its background knowledge and how it increases the potential risk of breaching the identities of blurred faces.en_US
dc.language.isoengen_US
dc.publisherSpringer natureen_US
dc.subjectAdversary modelen_US
dc.subjectBackground knowledgeen_US
dc.subjectDeep learning-assisted attacksen_US
dc.subjectFace obfuscationen_US
dc.subjectImage transformationen_US
dc.subjectPrivacy-preserving techniquesen_US
dc.titleA framework for evaluating image obfuscation under deep learning-assisted privacy attacksen_US
dc.typeJournal Articleen_US
dc.identifier.doi10.1007/s11042-023-14664-y-
dc.identifier.scopus2-s2.0-85160270905-
dc.identifier.urlhttps://api.elsevier.com/content/abstract/scopus_id/85160270905-
dc.contributor.affiliationDepartment of Mechatronics Engineeringen_US
dc.description.volume82en_US
dc.description.startpage42173en_US
dc.description.endpage42205en_US
dc.date.catalogued2023-06-06-
dc.description.statusPublisheden_US
dc.identifier.openURLhttps://link.springer.com/article/10.1007/s11042-023-14664-yen_US
dc.relation.ispartoftextMultimedia Tools and Applicationsen_US
crisitem.author.parentorgIssam Fares Faculty of Technology-
Appears in Collections:Department of Mechatronics Engineering
Show simple item record

SCOPUSTM   
Citations

1
checked on Jul 13, 2024

Record view(s)

61
checked on Jul 13, 2024

Google ScholarTM

Check

Dimensions Altmetric

Dimensions Altmetric


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