URAdv: A Novel Framework for Generating Ultra-Robust Adversarial Patches Against UAV Object Detection

In recent years, deep learning has been extensively deployed on unmanned aerial vehicles (UAVs), particularly for object detection. As the cornerstone of UAV-based object detection, deep neural networks are susceptible to adversarial attacks, with adversarial patches being a relatively straightforwa...

Full description

Saved in:
Bibliographic Details
Main Authors: Hailong Xi, Le Ru, Jiwei Tian, Bo Lu, Shiguang Hu, Wenfei Wang, Xiaohui Luan
Format: Article
Language:English
Published: MDPI AG 2025-02-01
Series:Mathematics
Subjects:
Online Access:https://www.mdpi.com/2227-7390/13/4/591
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849719784933949440
author Hailong Xi
Le Ru
Jiwei Tian
Bo Lu
Shiguang Hu
Wenfei Wang
Xiaohui Luan
author_facet Hailong Xi
Le Ru
Jiwei Tian
Bo Lu
Shiguang Hu
Wenfei Wang
Xiaohui Luan
author_sort Hailong Xi
collection DOAJ
description In recent years, deep learning has been extensively deployed on unmanned aerial vehicles (UAVs), particularly for object detection. As the cornerstone of UAV-based object detection, deep neural networks are susceptible to adversarial attacks, with adversarial patches being a relatively straightforward method to implement. However, current research on adversarial patches, especially those targeting UAV object detection, is limited. This scarcity is notable given the complex and dynamically changing environment inherent in UAV image acquisition, which necessitates the development of more robust adversarial patches to achieve effective attacks. To address the challenge of adversarial attacks in UAV high-altitude reconnaissance, this paper presents a robust adversarial patch generation framework. Firstly, the dataset is reconstructed by considering various environmental factors that UAVs may encounter during image collection, and the influences of reflections and shadows during photography are integrated into patch training. Additionally, a nested optimization method is employed to enhance the continuity of attacks across different altitudes. Experimental results demonstrate that the adversarial patches generated by the proposed method exhibit greater robustness in complex environments and have better transferability among similar models.
format Article
id doaj-art-67564dd1de9647f2a5679ec3e8fdd50d
institution DOAJ
issn 2227-7390
language English
publishDate 2025-02-01
publisher MDPI AG
record_format Article
series Mathematics
spelling doaj-art-67564dd1de9647f2a5679ec3e8fdd50d2025-08-20T03:12:05ZengMDPI AGMathematics2227-73902025-02-0113459110.3390/math13040591URAdv: A Novel Framework for Generating Ultra-Robust Adversarial Patches Against UAV Object DetectionHailong Xi0Le Ru1Jiwei Tian2Bo Lu3Shiguang Hu4Wenfei Wang5Xiaohui Luan6Equipment Management and Unmanned Aerial Vehicle Engineering School, Air Force Engineering University, Xi’an 710043, ChinaEquipment Management and Unmanned Aerial Vehicle Engineering School, Air Force Engineering University, Xi’an 710043, ChinaAir Traffic Control and Navigation School, Air Force Engineering University, Xi’an 710043, ChinaEquipment Management and Unmanned Aerial Vehicle Engineering School, Air Force Engineering University, Xi’an 710043, ChinaEquipment Management and Unmanned Aerial Vehicle Engineering School, Air Force Engineering University, Xi’an 710043, ChinaEquipment Management and Unmanned Aerial Vehicle Engineering School, Air Force Engineering University, Xi’an 710043, ChinaChina Academy of Space Technology (Xi’an), Xi’an 710000, ChinaIn recent years, deep learning has been extensively deployed on unmanned aerial vehicles (UAVs), particularly for object detection. As the cornerstone of UAV-based object detection, deep neural networks are susceptible to adversarial attacks, with adversarial patches being a relatively straightforward method to implement. However, current research on adversarial patches, especially those targeting UAV object detection, is limited. This scarcity is notable given the complex and dynamically changing environment inherent in UAV image acquisition, which necessitates the development of more robust adversarial patches to achieve effective attacks. To address the challenge of adversarial attacks in UAV high-altitude reconnaissance, this paper presents a robust adversarial patch generation framework. Firstly, the dataset is reconstructed by considering various environmental factors that UAVs may encounter during image collection, and the influences of reflections and shadows during photography are integrated into patch training. Additionally, a nested optimization method is employed to enhance the continuity of attacks across different altitudes. Experimental results demonstrate that the adversarial patches generated by the proposed method exhibit greater robustness in complex environments and have better transferability among similar models.https://www.mdpi.com/2227-7390/13/4/591unmanned aerial vehiclesobject detectionadversarial attackadversarial patch
spellingShingle Hailong Xi
Le Ru
Jiwei Tian
Bo Lu
Shiguang Hu
Wenfei Wang
Xiaohui Luan
URAdv: A Novel Framework for Generating Ultra-Robust Adversarial Patches Against UAV Object Detection
Mathematics
unmanned aerial vehicles
object detection
adversarial attack
adversarial patch
title URAdv: A Novel Framework for Generating Ultra-Robust Adversarial Patches Against UAV Object Detection
title_full URAdv: A Novel Framework for Generating Ultra-Robust Adversarial Patches Against UAV Object Detection
title_fullStr URAdv: A Novel Framework for Generating Ultra-Robust Adversarial Patches Against UAV Object Detection
title_full_unstemmed URAdv: A Novel Framework for Generating Ultra-Robust Adversarial Patches Against UAV Object Detection
title_short URAdv: A Novel Framework for Generating Ultra-Robust Adversarial Patches Against UAV Object Detection
title_sort uradv a novel framework for generating ultra robust adversarial patches against uav object detection
topic unmanned aerial vehicles
object detection
adversarial attack
adversarial patch
url https://www.mdpi.com/2227-7390/13/4/591
work_keys_str_mv AT hailongxi uradvanovelframeworkforgeneratingultrarobustadversarialpatchesagainstuavobjectdetection
AT leru uradvanovelframeworkforgeneratingultrarobustadversarialpatchesagainstuavobjectdetection
AT jiweitian uradvanovelframeworkforgeneratingultrarobustadversarialpatchesagainstuavobjectdetection
AT bolu uradvanovelframeworkforgeneratingultrarobustadversarialpatchesagainstuavobjectdetection
AT shiguanghu uradvanovelframeworkforgeneratingultrarobustadversarialpatchesagainstuavobjectdetection
AT wenfeiwang uradvanovelframeworkforgeneratingultrarobustadversarialpatchesagainstuavobjectdetection
AT xiaohuiluan uradvanovelframeworkforgeneratingultrarobustadversarialpatchesagainstuavobjectdetection