A Study of Adversarial Attacks and Detection on Deep Learning-Based Plant Disease Identification

Transfer learning using pre-trained deep neural networks (DNNs) has been widely used for plant disease identification recently. However, pre-trained DNNs are susceptible to adversarial attacks which generate adversarial samples causing DNN models to make wrong predictions. Successful adversarial att...

Full description

Saved in:
Bibliographic Details
Published inApplied sciences Vol. 11; no. 4; p. 1878
Main Authors Luo, Zhirui, Li, Qingqing, Zheng, Jun
Format Journal Article
LanguageEnglish
Published Basel MDPI AG 01.02.2021
Subjects
adversarial attacks
adversarial sample detection
Agricultural production
Algorithms
Automation
Datasets
Deep learning
Identification
Machine learning
Neural networks
Pathogens
plant disease identification
Plant diseases
Subject specialists
Unmanned aerial vehicles
white-box attacks
Online AccessGet full text

Cover

More Information
Summary:Transfer learning using pre-trained deep neural networks (DNNs) has been widely used for plant disease identification recently. However, pre-trained DNNs are susceptible to adversarial attacks which generate adversarial samples causing DNN models to make wrong predictions. Successful adversarial attacks on deep learning (DL)-based plant disease identification systems could result in a significant delay of treatments and huge economic losses. This paper is the first attempt to study adversarial attacks and detection on DL-based plant disease identification. Our results show that adversarial attacks with a small number of perturbations can dramatically degrade the performance of DNN models for plant disease identification. We also find that adversarial attacks can be effectively defended by using adversarial sample detection with an appropriate choice of features. Our work will serve as a basis for developing more robust DNN models for plant disease identification and guiding the defense against adversarial attacks.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2076-3417
2076-3417
DOI:10.3390/app11041878