OD-SHIELD: Convolutional Autoencoder-Based Defense Against Adversarial Patch Attacks in Object Detection

In the evolving landscape of deep neural network security, adversarial patch attacks present a serious challenge for object detection systems. We introduce <sc>OD-Shield</sc>, a novel defense approach that employs a convolutional autoencoder framework to detect and remove anomalous regio...

Full description

Saved in:
Bibliographic Details
Main Authors: Byeongchan Kim, Heemin Kim, Minjung Kang, Hyunjee Nam, Sunghwan Park, Jaewoo Lee, Il-Youp Kwak
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11021559/
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In the evolving landscape of deep neural network security, adversarial patch attacks present a serious challenge for object detection systems. We introduce <sc>OD-Shield</sc>, a novel defense approach that employs a convolutional autoencoder framework to detect and remove anomalous regions in patched images, subsequently reconstructing these regions to mitigate adversarial effects. The reconstructed images are then processed by the object detector, thereby restoring reliable performance under diverse attack scenarios. Distinctly model-agnostic, <sc>OD-Shield</sc> operates as a pre-processing step and can be applied to a wide range of tasks&#x2014;including image classification and object detection&#x2014;without compromising the fidelity of the original image. Experiments on benchmark datasets (COCO, Visdrone, and Argoverse) reveal that <sc>OD-Shield</sc> outperforms existing defenses by 13%&#x2013;47% on COCO, highlighting its effectiveness in addressing critical security vulnerabilities. This work not only tackles the immediate threat of adversarial patches but also lays the foundation for future research into adaptive, resilient defense mechanisms that keep pace with evolving adversarial tactics.
ISSN:2169-3536