收藏切换
Perceptual Similarity-Based Multi-Objective Optimization for Stealthy Image Backdoor Attack
收藏切换
Suxia Zhu, Jinyin Wang, Guanglu Sun*
Journal of Computer Research and Development | 2024, 61(5) : 1182 - 1192
Less
收藏切换
Journal of Computer Research and Development | 2024, 61(5): 1182-1192
• Special Issue on Large Models and Safety •
Perceptual Similarity-Based Multi-Objective Optimization for Stealthy Image Backdoor Attack
Full
Suxia Zhu, Jinyin Wang, Guanglu Sun*
Affiliations
  • School of Computer Science and Technology, Harbin University of Science and Technology, Harbin 150080
  • Heilongjiang Key Laboratory of Intelligent Information Processing and Application (Harbin University of Science and Technology), Harbin 150080
  • born in 1978. PhD, professor, PhD supervisor. Member of CCF. Her main research interests include privacy and security, IoT, and parallel computing

    born in 1998. Master candidate. His main research interests include machine learning security and privacy preserving

    born in 1979. PhD, professor, PhD supervisor. Distinguished member of CCF. His main research interests include artificial intelligence, network and information security, and intelligent information processing

About Author:

born in 1978. PhD, professor, PhD supervisor. Member of CCF. Her main research interests include privacy and security, IoT, and parallel computing

born in 1998. Master candidate. His main research interests include machine learning security and privacy preserving

born in 1979. PhD, professor, PhD supervisor. Distinguished member of CCF. His main research interests include artificial intelligence, network and information security, and intelligent information processing

Published: 2024-05-01 doi: 10.7544/issn1000-1239.202330521
Outline
收藏切换

Deep learning models are vulnerable to backdoor attacks and behave normally when processing clean data, but they will exhibit malicious behavior when processing toxic samples with trigger patterns. However, most backdoor attacks currently produce backdoor images that are easily perceived by the human eye, resulting in insufficient stealthiness of backdoor attacks. Therefore, a multi-objective optimized covert image backdoor attack method based on perceptual similarity is proposed. Firstly, the visual difference between the backdoor image and the original image is reduced using a perceptual similarity loss function. Secondly, a multi-objective optimization method is used to solve the problem of inter-task conflict on the poisoning model, thus ensuring stable performance of the model after poisoning. Finally, a two-stage training method is adopted to automate the generation of trigger patterns and improve the training efficiency. The final experimental results show that it is difficult for human eye to distinguish the generated backdoor image from the original image without any degradation in clean accuracy. Meanwhile, the backdoor attack is successfully performed on the target classification model, and the attack success rate reaches 100% for all experimental datasets under the all-to-one attack strategy. Compared with other steganographic backdoor attack methods, our method has better stealthiness.

backdoor attack  /  covert backdoor  /  poisoning attack  /  deep learning  /  model security
朱素霞, 王金印, 孙广路. 基于感知相似性的多目标优化隐蔽图像后门攻击. 计算机研究与发展, 2024 , 61 (5) : 1182 -1192 . DOI: 10.7544/issn1000-1239.202330521
Suxia Zhu, Jinyin Wang, Guanglu Sun. Perceptual Similarity-Based Multi-Objective Optimization for Stealthy Image Backdoor Attack[J]. Journal of Computer Research and Development, 2024 , 61 (5) : 1182 -1192 . DOI: 10.7544/issn1000-1239.202330521
Year 2024 volume 61 Issue 5
80
40
Cite this Article
BibTeX
Article Info
doi: 10.7544/issn1000-1239.202330521
  • Receive Date:2023-06-19
  • Online Date:2026-11-17
  • Published:2024-05-01
Article Data
Affiliations
History
  • Received:2023-06-19
  • Revised:2024-01-25
Affiliations
    School of Computer Science and Technology, Harbin University of Science and Technology, Harbin 150080
    Heilongjiang Key Laboratory of Intelligent Information Processing and Application (Harbin University of Science and Technology), Harbin 150080
References
Share
https://castjournals.cast.org.cn/joweb/jsjyjyfz/EN/10.7544/issn1000-1239.202330521
Share to
QR

Scan QR to access full text

Cite this article
BibTeX
Citations
表12种不同金属材料的力学参数
科
Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)
属
Genus
种数
Number of
species
占总种数比例
Percentage of total
species (%)
鹅膏菌科Amanitaceae 2 11 5.26 鹅膏菌属 Amanita 10 4.78
小菇科 Mycenaceae 2 12 5.74 丝盖伞属 Inocybe 5 2.39
多孔菌科 Polyporaceae 8 14 6.70 蜡蘑属 Laccaria 5 2.39
红菇科 Russulaceae 3 23 11.00 小皮伞属 Marasmius 6 2.87
小菇属 Mycena 11 5.26
光柄菇属 Pluteus 5 2.39
红菇属 Russula 17 8.13
栓菌属 Trametes 5 2.39
关闭全屏
  • BibTeX
  • EndNote
  • RefWorks
  • TxT