Sigma-UAP:一种针对深度神经网络的隐形半通用对抗性攻击

Feiyang Qin, Wenqi Na, Song Gao, Shaowen Yao
{"title":"Sigma-UAP:一种针对深度神经网络的隐形半通用对抗性攻击","authors":"Feiyang Qin, Wenqi Na, Song Gao, Shaowen Yao","doi":"10.1109/prmvia58252.2023.00012","DOIUrl":null,"url":null,"abstract":"Although deep neural networks (DNNs) have achieved exceptional performance, they are shown to be fragile to universal adversarial perturbations (UAP), which can be applied to any images to fool a well-trained DNN. Several methods have been proposed to design universal perturbations. However, these methods often leave visible traces in natural images. In this paper, we propose Sigma-UAP, a semi-universal adversarial attack, to enhance the quasi-imperceptibility of universal adversarial perturbations, in which the Sigma-map algorithm is leveraged to hide perturbations by identifying the low-frequency region of the image and eliminating the perturbations in that region. Then, we use a simple matrix calculation to augment the perturbation in the high-frequency region to ensure the attack effectiveness of the perturbation. The extensive empirical experiments show that, compared with the state-of-the-art universal adversarial attacks, Sigma-UAP method obtains excellent attack capabilities in visual effect and attack success rate.","PeriodicalId":221346,"journal":{"name":"2023 International Conference on Pattern Recognition, Machine Vision and Intelligent Algorithms (PRMVIA)","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Sigma-UAP: An Invisible Semi-Universal Adversarial Attack Against Deep Neural Networks\",\"authors\":\"Feiyang Qin, Wenqi Na, Song Gao, Shaowen Yao\",\"doi\":\"10.1109/prmvia58252.2023.00012\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Although deep neural networks (DNNs) have achieved exceptional performance, they are shown to be fragile to universal adversarial perturbations (UAP), which can be applied to any images to fool a well-trained DNN. Several methods have been proposed to design universal perturbations. However, these methods often leave visible traces in natural images. In this paper, we propose Sigma-UAP, a semi-universal adversarial attack, to enhance the quasi-imperceptibility of universal adversarial perturbations, in which the Sigma-map algorithm is leveraged to hide perturbations by identifying the low-frequency region of the image and eliminating the perturbations in that region. Then, we use a simple matrix calculation to augment the perturbation in the high-frequency region to ensure the attack effectiveness of the perturbation. The extensive empirical experiments show that, compared with the state-of-the-art universal adversarial attacks, Sigma-UAP method obtains excellent attack capabilities in visual effect and attack success rate.\",\"PeriodicalId\":221346,\"journal\":{\"name\":\"2023 International Conference on Pattern Recognition, Machine Vision and Intelligent Algorithms (PRMVIA)\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2023-03-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2023 International Conference on Pattern Recognition, Machine Vision and Intelligent Algorithms (PRMVIA)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/prmvia58252.2023.00012\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2023 International Conference on Pattern Recognition, Machine Vision and Intelligent Algorithms (PRMVIA)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/prmvia58252.2023.00012","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

摘要

尽管深度神经网络(DNN)已经取得了卓越的表现,但它们在普遍对抗性扰动(UAP)面前很脆弱,这可以应用于任何图像来欺骗训练良好的DNN。已经提出了几种设计通用摄动的方法。然而,这些方法往往会在自然图像中留下可见的痕迹。在本文中,我们提出了一种半通用对抗性攻击Sigma-UAP,以增强通用对抗性扰动的准不可感知性,其中利用Sigma-map算法通过识别图像的低频区域并消除该区域的扰动来隐藏扰动。然后,我们使用简单的矩阵计算来增加高频区域的扰动,以确保扰动的攻击有效性。大量的实证实验表明,与目前最先进的通用对抗性攻击相比,Sigma-UAP方法在视觉效果和攻击成功率方面都具有优异的攻击能力。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Sigma-UAP: An Invisible Semi-Universal Adversarial Attack Against Deep Neural Networks
Although deep neural networks (DNNs) have achieved exceptional performance, they are shown to be fragile to universal adversarial perturbations (UAP), which can be applied to any images to fool a well-trained DNN. Several methods have been proposed to design universal perturbations. However, these methods often leave visible traces in natural images. In this paper, we propose Sigma-UAP, a semi-universal adversarial attack, to enhance the quasi-imperceptibility of universal adversarial perturbations, in which the Sigma-map algorithm is leveraged to hide perturbations by identifying the low-frequency region of the image and eliminating the perturbations in that region. Then, we use a simple matrix calculation to augment the perturbation in the high-frequency region to ensure the attack effectiveness of the perturbation. The extensive empirical experiments show that, compared with the state-of-the-art universal adversarial attacks, Sigma-UAP method obtains excellent attack capabilities in visual effect and attack success rate.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信