欢迎访问《汽车安全与节能学报》，
汽车安全与节能学报

汽车安全与节能学报 ›› 2025, Vol. 16 ›› Issue (1): 50-56.DOI: 10.3969/j.issn.1674-8484.2025.01.005

• 汽车安全 • 上一篇    下一篇

基于视觉的汽车检测系统物理对抗攻击

刘育秋(), 唐亮*(), 王宁珍   

  1. 北京林业大学 工学院，北京 100083，中国
  • 收稿日期:2024-03-31 修回日期:2024-07-18 出版日期:2025-02-28 发布日期:2025-03-04
  • 通讯作者: * 唐亮，教授。E-mail：happyliang@bjfu.edu.cn。
  • 作者简介:刘育秋（1999—），女（汉），山西，硕士研究生。E-mail：yuqiu99@bjfu.edu.cn。
  • 基金资助:
    汽车零部件先进制造技术教育部重点实验室开放课题基金(2021KLMT05)

Physical adversarial attack on vehicle detection systems

LIU Yuqiu(), TANG Liang*(), WANG Ningzhen   

  1. School of Technology, Bejing Forestry University, Beijing 10083, China
  • Received:2024-03-31 Revised:2024-07-18 Online:2025-02-28 Published:2025-03-04

RichHTML

4

PDF

83

摘要：

为保护个人车辆的隐私信息，实现汽车伪装，对基于视觉的汽车检测系统进行物理对抗攻击。在3D对抗攻击框架的基础上，对攻击算法的目标函数进行优化设计，提升3D对抗纹理在多视角多场景设定下的攻击效果，设计针对天气变化的带权分层颜色映射网络，使对抗纹理能够在训练流程中表达在天气参数下的颜色反应，从而进一步增强攻击在物理世界实现的鲁棒性。进行了数字和物理实验。结果表明：基于视觉的汽车检测系统在本算法攻击下检测的平均召回率降低了49.4%。从而，本算法优化出的对抗纹理具备的物理世界可实现性，能够在物理世界实现使检测系统的准确率下降至少38.7%。

关键词: 汽车检测系统, 物理对抗攻击, 深度学习, 计算机视觉

Abstract:

A camouflage method was developed for conducting physical adversarial attacks on vision-based vehicle detection systems to protect the privacy of personal vehicles. An objective function of the attack algorithm was optimized based on a 3D adversarial attack framework to enhance the effectiveness of 3D adversarial textures under multi-view and multi-scene settings. A weather-adaptive weighted hierarchical color mapping network was designed to enable adversarial textures to respond to weather parameters during the training process, further improving the robustness of physical-world attacks. Digital and physical experiments were conducted. The results show that the proposed algorithm reduces the average recall rate of detection by 49.4%. Therefore, the optimized adversarial textures demonstrate physical-world feasibility, achieving at least a 38.7% reduction in detection accuracy in real-world scenarios.

Key words: vehicle detection systems, physical adversarial texture attacks, deep learning, computer vision

中图分类号: