CGAN-IRB: a novel data augmentation method for apple leaf diseases
Xinbin Yuan¹, Cong Yu¹, Bin Liu¹, Henan Sun¹, Xianyu Zhu¹
¹College of Information Engineering, Northwest A&F University, Yangling, China
Published in In the proceedings of 2021 IEEE 45th Annual Computers, Software, and Applications Conference (COMPSAC), 2021
Keywords: CGAN, data augmentation, apple leaf diseases, IRB
Abstract
This paper presents CGAN-IRB, a novel data augmentation method for apple leaf diseases using Conditional Generative Adversarial Networks with Inverted Residual Block. The proposed method addresses the challenge of limited training data in agricultural disease detection by generating synthetic samples that maintain the characteristics of real apple leaf diseases. Experimental results demonstrate the effectiveness of the proposed approach in improving model performance for apple leaf disease classification tasks.
Recommended citation: Xinbin Yuan, Cong Yu, Bin Liu, Henan Sun, Xianyu Zhu, "CGAN-IRB: a novel data augmentation method for apple leaf diseases." In the proceedings of 2021 IEEE 45th Annual Computers, Software, and Applications Conference (COMPSAC), 2021.
@inproceedings{yuan2021cgan,
title={CGAN-IRB: a novel data augmentation method for apple leaf diseases},
author={Yuan, Xinbin and Yu, Cong and Liu, Bin and Sun, Henan and Zhu, Xianyu},
booktitle={In the proceedings of 2021 IEEE 45th Annual Computers, Software, and Applications Conference (COMPSAC)},
year={2021},
publisher={IEEE}
}