Adversarial Attacks Defense Method Based on Multiple Filtering and Image Rotation

Feng Li, Xuehui Du, Liu Zhang and Ahmed Farouk

Discrete Dynamics in Nature and Society, 2021, vol. 2021, 1-11

Abstract: Adversarial examples in an image classification task cause neural networks to predict incorrect class labels with high confidence. Many applications related to image classification, such as self-driving and facial recognition, have been seriously threatened by adversarial attacks. One class of the existing defense methods is the preprocessing-based defense which transforms the inputs before feeding them to the system. These methods are independent of the classification models and have excellent defensive effects under oblivious attacks. An image filtering method is often used to evaluate the robustness of adversarial examples. However, filtering induces the loss of valuable features that reduce the classification accuracy and weakens the adversarial perturbation. Furthermore, the fixed filtering parameters cannot effectively defend against the adversarial attack. This paper proposes a novel defense method based on different filter parameters and randomly rotated filtered images. The output classification probabilities are statistically averaged, which keeps the classification accuracy while removing the perturbation. Experimental results show that the proposed method improves the defense capability of various models against diverse kinds of oblivious adversarial attacks. Under the adaptive attack, the transferability of the adversarial examples among different models is significantly reduced.

Date: 2021
References: Add references at CitEc
Citations:

Downloads: (external link)
http://downloads.hindawi.com/journals/ddns/2021/6124895.pdf (application/pdf)
http://downloads.hindawi.com/journals/ddns/2021/6124895.xml (application/xml)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:hin:jnddns:6124895

DOI: 10.1155/2021/6124895

Access Statistics for this article

More articles in Discrete Dynamics in Nature and Society from Hindawi
Bibliographic data for series maintained by Mohamed Abdelhakeem ().