Model Agnostic Defense against Adversarial Patch Attacks on Object Detection in Unmanned Aerial Vehicles

TL;DR

This paper introduces a lightweight, model-agnostic defense mechanism against adversarial patch attacks on UAV-based object detection, effectively neutralizing patches without retraining and maintaining high detection reliability.

Contribution

A novel occlusion removal-based defense method that is model-agnostic, does not require adversarial patches during training, and is suitable for real-time UAV applications.

Findings

Significantly reduces attack success ratio in digital and physical tests.

Maintains low computational costs suitable for UAV deployment.

Enhances the robustness of UAV object detection systems.

Abstract

Object detection forms a key component in Unmanned Aerial Vehicles (UAVs) for completing high-level tasks that depend on the awareness of objects on the ground from an aerial perspective. In that scenario, adversarial patch attacks on an onboard object detector can severely impair the performance of upstream tasks. This paper proposes a novel model-agnostic defense mechanism against the threat of adversarial patch attacks in the context of UAV-based object detection. We formulate adversarial patch defense as an occlusion removal task. The proposed defense method can neutralize adversarial patches located on objects of interest, without exposure to adversarial patches during training. Our lightweight single-stage defense approach allows us to maintain a model-agnostic nature, that once deployed does not require to be updated in response to changes in the object detection pipeline. The evaluations in digital and physical domains show the feasibility of our method for deployment in UAV object detection pipelines, by significantly decreasing the Attack Success Ratio without incurring significant processing costs. As a result, the proposed defense solution can improve the reliability of object detection for UAVs.