Attacks based on malicious perturbations on image processing systems and defense methods against them

Attacks based on malicious perturbations on image processing systems and defense methods against them

Systems implementing artificial intelligence technologies have become widespread due to their effectiveness in solving various applied tasks including computer vision. Image processing through neural networks is also used in securitycritical systems. At the same time, the use of artificial intellige...

Full description

Bibliographic Details
Main Authors: Dmitry A. Esipov, Abdulhamid Y. Buchaev, Akylzhan Kerimbay, Yana V. Puzikova, Semen K. Saidumarov, Nikita S. Sulimenko, Ilya Yu. Popov, Nikolay S. Karmanovskiy
Format: Article
Language:English
Published: Saint Petersburg National Research University of Information Technologies, Mechanics and Optics (ITMO University) 2023-08-01
Series:Naučno-tehničeskij Vestnik Informacionnyh Tehnologij, Mehaniki i Optiki
Subjects:
artificial intelligence
artificial neural network
image processing
adversarial attack
backdoor embedding
adversarial perturbation
adversarial learning
defense distillation
feature squeezing
certified defense
data preprocessing
Online Access:https://ntv.ifmo.ru/file/article/22193.pdf

Internet

https://ntv.ifmo.ru/file/article/22193.pdf

Similar Items