基于改进CycleGAN的小样本玉米病害图像扩充方法

李艳玲; 张博翔; 李飞涛; Bacao Fernando; 司海平; 陈丽娜

doi:10.13300/j.cnki.hnlkxb.2025.05.020

您当前的位置：

首页 >

文章列表页 >

基于改进CycleGAN的小样本玉米病害图像扩充方法

植物保护 | 更新时间：2026-11-05

- 基于改进CycleGAN的小样本玉米病害图像扩充方法
- Improved CycleGAN-based method for augmenting images of few-shot maize diseases
- 华中农业大学学报 2025年44卷第5期页码：198-207
- 作者机构：
  
  1.河南农业大学信息与管理科学学院，郑州 450002
  2.新里斯本大学信息管理学院，里斯本 1070⁃312
  3.商丘师范学院计算机与信息技术学院，商丘 476000
- 作者简介：
  
  李艳玲，E-mail：lyl_lingling@163.com
  司海平，E-mail：pingsss@126.com
  陈丽娜，E-mail：dxn1126@163.com
- 基金信息：
  
  河南省科技攻关项目(252102520037);河南省重点研发专项(251111211300;231111110100;231111211300);河南省杰出外籍科学家工作室项目(GZS2024006);河南省教育厅高等学校重点项目(25A520044)
- DOI：10.13300/j.cnki.hnlkxb.2025.05.020
  中图分类号： S513;TP391.1
- 收稿：2025-03-22，
  
  纸质出版：2025-09-15
- 稿件说明：
移动端阅览
李艳玲，张博翔，李飞涛，等.基于改进CycleGAN的小样本玉米病害图像扩充方法［J］.华中农业大学学报，2025，44（5）：198‐207.

LI Yanling,ZHANG Boxiang,LI Feitao,et al.Improved CycleGAN-based method for augmenting images of few-shot maize diseases[J].Journal of Huazhong Agricultural University,2025,44(05):198-207.
李艳玲，张博翔，李飞涛，等.基于改进CycleGAN的小样本玉米病害图像扩充方法［J］.华中农业大学学报，2025，44（5）：198‐207. DOI： 10.13300/j.cnki.hnlkxb.2025.05.020.

LI Yanling,ZHANG Boxiang,LI Feitao,et al.Improved CycleGAN-based method for augmenting images of few-shot maize diseases[J].Journal of Huazhong Agricultural University,2025,44(05):198-207. DOI： 10.13300/j.cnki.hnlkxb.2025.05.020.

摘要

针对玉米病害图像识别任务存在数据集获取困难、样本不足及不同类别病害样本不均衡等问题，设计一种基于改进CycleGAN（cycle-consistent adversarial networks）的图像数据增强方法。首先，使用较小感受野的卷积核优化CycleGAN网络结构，生成高质量样本图像，减少过拟合现象发生；其次，将SE（squeeze-excitation）注意力机制嵌入到生成器的残差模块中，增强CycleGAN对病害特征的提取能力，使网络更准确地捕捉小目标病害或域间差异不明显的特征。结果显示，改进后的CycleGAN相较于原始CycleGAN、DCGAN、DCGAN+和WGAN算法，生成病害图像的FID分数分别降低了43.33、32.67、24.24和19.72，GAN-train与GAN-test相较于原始CycleGAN提升了3.13、4.25百分点；采用改进的CycleGAN图像扩充方法构建玉米病害数据集，基于该数据集的玉米叶片病害识别模型准确率在3种网络架构上均得到有效提升：AlexNet提升3.90百分点，VGGNet提升4.41百分点，ResNet提升3.44百分点，在ResNet网络架构上与传统数据增强算法相比病害识别率提升5.79百分点。结果表明，改进的CycleGAN网络有效解决了玉米病害图像数据集匮乏的问题。

Abstract

An improved cycle-consistent adversarial networks （CycleGAN）-based method for augmenting data of maize leaf disease images was designed to solve the difficulties in obtaining image dataset of recognizing maize diseases，insufficient samples，and imbalanced samples across different categories of diseases.Convolutional kernels with smaller receptive fields were used to optimize the structure of the CycleGAN network and generate sample images with high-quality and reduce the occurrence of overfitting.The SE （squeeze-excitation） attention mechanism was embedded into the residual module of the generator to enhance the ability of CycleGAN to extract disease features and allow the network to more accurately capture diseases with small target or features with subtle inter-domain differences.The results showed that the improved CycleGAN reduced the frechet inception distance （FID） value of generated disease images by 43.33，32.67，24.24，and 19.72 compared with the original CycleGAN，DCGAN，DCGAN+，and WGAN algorithms. GAN-train and GAN-test increased by 3.13 and 4.25 percentage points compared to that in the original CycleGAN.The improved CycleGAN-based method for augmenting data was used to construct a dataset of maize diseases，and the accuracy of recognizing the maize leaf diseases based on this dataset was significantly improved on three network architectures.AlexNet，VGGNet，and ResNet increased by 3.90，4.41，and 3.44 percentage points，respectively.Compared with traditional data augmentation algorithms，the disease recognition rate of ResNet network architecture increased by 5.79 percentage points.It is indicated that the improved CycleGAN network effectively solves the problem of insufficient image dataset of maize diseases.

关键词

Keywords

references

ROUF SHAH T，PRASAD K，KUMAR P.Maize，a potential source of human nutrition and health：a review［J/OL］.Cogent food & agriculture，2016，2（1）：1166995［2025-03-22］.https：//doi.org/10.1080/23311932.2016.1166995.

ESMAEILINEZHAD A，ABBASI H.Physicochemical，thermal and microstructural properties of new developed bigels through Zein and Persian gum biopolymers［J］.Journal of food measurement and characterization，2025，19（3）：1693-1706.

KUSUMO B S，HERYANA A，MAHENDRA O，et al.Machine learning-based for automatic detection of corn-plant diseases using image processing［C］//2018 International Conference on Computer，Control，Informatics and its Applications （IC3INA）. November 1-2，2018 ，Tangerang，Indonesia.Tangerang：IEEE，2018：93-97.

ZHANG X H，QIAO Y，MENG F F，et al.Identification of maize leaf diseases using improved deep convolutional neural networks［J］.IEEE access，2018，6：30370-30377.

SHARMA V，TRIPATHI A K，DAGA P，et al.ClGanNet：a novel method for maize leaf disease identification using ClGan and deep CNN［J/OL］.Signal processing：image communication，2024，120：117074［2025-03-22］.https：//doi.org/10.1016/j.image.2023.117074.

CARRARO A，SAURIO G，MARINELLO F.Towards rigorous dataset quality standards for deep learning tasks in precision agriculture：a case study exploration［J/OL］.Smart agricultural technology，2025，10：100721［2025-03-22］.https：//doi.org/10.1016/j.atech.2024.100721.

ALOM M Z，TAHA T M，YAKOPCIC C，et al.The history began from alexnet：a comprehensive survey on deep learning approaches［DB/OL］.arXiv，2018：1803.01164［2025-03-22］.https：//doi.org/10.48550/arXiv.1803.01164.

SANDLER M，HOWARD A，ZHU M L，et al.MobileNetV2：inverted residuals and linear bottlenecks［C］//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. June 18-23，2018 ，Salt Lake City，USA.Salt Lake City：IEEE，2018：4510-4520.

SIMONYAN K，ZISSERMAN A.Very deep convolutional networks for large-scale image recognition［DB/OL］.arXiv，2014：1409.1556［2025-03-22］.https：//doi.org/10.48550/arXiv.1409.1556.

AL-QIZWINI M，BARJASTEH I，AL-QASSAB H，et al.Deep learning algorithm for autonomous driving using GoogLeNet［C］//2017 IEEE Intelligent Vehicles Symposium （Ⅳ）. June 11-14，2017 ，Los Angeles，CA，USA.Los Angeles，IEEE，2017：89-96.

HUANG G，LIU S，VAN DER MAATEN L，et al.Condensenet：An efficient densenet using learned group convolutions［DB/OL］.arXiv，2018：1711.09224［2025-03-22］.https：//doi.org/10.48550/arXiv.1711.09224.

HE K M，ZHANG X Y，REN S Q，et al.Deep residual learning for image recognition［C］//2016 IEEE Conference on Computer Vision and Pattern Recognition （CVPR）. June 27-30，2016 ，Las Vegas，NV，USA.Las Vegas：IEEE，2016：770-778.

BAO W X，HUANG X F，HU G S，et al.Identification of maize leaf diseases using improved convolutional neural network［J］.Transactions of the CSAE，2021，37（6）：160-167.

BANSAL S，KUMAR A.Automatic deep neural network hyper-parameter optimization for maize disease detection［J/OL］.IOP conference series：materials science and engineering，2021，1022（1）：012089［2025-03-22］.https：//doi.org/10.1088/1757-899X/1022/1/012089.

马铁民，曲浩，高雅，等.基于超分辨率生成对抗网络的玉米病害分类识别方法［J］.农业机械学报，2024，55（11）：49-56.MA T M，QU H，GAO Y，et al.Maize disease classification and recognition method based on super-resolution generative adversarial networks［J］.Transactions of the CSAM，2024，55（11）：49-56 （in Chinese with English abstract） .

武芳，王芹，毛洪贲.小样本在线学习数据处理及学生成绩预测方法研究［J］.医药高职教育与现代护理，2024，7（6）：502-508.WU F，WANG Q，MAO H B.Research on small sample online learning data processing and student performance prediction methods［J］.Medical higher vocational ed ucation and modern nursing，2024，7（6）：502-508 （in Chinese with English abstract）

STEPHEN A，PUNITHA A，CHANDRASEKAR A.Optimal deep generative adversarial network and convolutional neural network for rice leaf disease prediction［J］.The visual computer，2024，40（2）：919-936.

王志强，于雪莹，杨晓婧，等.基于WGAN和MCA-MobileNet的番茄叶片病害识别［J］.农业机械学报，2023，54（5）：244-252.WANG Z Q，YU X Y，YANG X J，et al.Tomato leaf diseases recognition based on WGAN and MCA-MobileNet［J］.Transactions of the CSAM，2023，54（5）：244-252 （in Chinese with English abstract） .

曾明昭，高会议，万莉.基于生成对抗网络的葡萄叶片图像数据增强方法［J］.仪表技术，2021（5）：41-44.ZENG M Z，GAO H Y，WAN L.Data augmentation of grape leaf diseases based on generative adversarial network［J］.Instrumentation technology，2021（5）：41-44 （in Chinese with English abstract） .

CHEN Y P，WU Q F.Grape leaf disease identification with sparse data via generative adversarial networks and convolutional neural networks［J］.Precision agriculture，2023，24（1）：235-253.

满超，饶元，张敬尧，等.基于改进Inception网络的复杂环境下小样本黄瓜叶片病害识别［J］.华中农业大学学报，2023，42（3）：152-160.MAN C，RAO Y，ZHANG J Y，et al.Recognition of cucumber leaf disease with small samples in complex environment based on improved Inception network［J］.Journal of Huazhong Agricultural University，2023，42（3）：152-160 （in Chinese with English abstract） .

叶阳，沈冰雁，沈毓琦.基于生成对抗网络的抗阴影树木检测方法［J］.农业工程学报，2021，37（10）：118-126.YE Y，SHEN B Y，SHEN Y Q.Research on anti-shadow tree detection method based on generative adversarial network［J］.Transactions of the CSAE，2021，37（10）：118-126 （in Chinese with English abstract） .

LIANG X D，HU Z T，ZHANG H，et al.Recurrent topic-transition GAN for visual paragraph generation［C］//2017 IEEE International Conference on Computer Vision （ICCV）. October 22-29，2017 ，Venice，Italy.Venice：IEEE，2017：3382-3391.

韩翔，李玉强，高昂，等.基于改进生成对抗网络的甜樱桃数据增强方法［J］.农业机械学报，2024，55（10）：252-262.HAN X，LI Y Q，GAO A，et al.Data augmentation method for sweet cherries based on improved generative adversarial network［J］.Transactions of the CSAM，2024，55（10）：252-262 （in Chinese with English abstract） .

XU B，ZHOU D，LI W J.Image enhancement algorithm based on GAN neural network［J］.IEEE access，2022，10：36766-36777.

BERTOGLIO R，MAZZUCCHELLI A，CATALANO N，et al.A comparative study of Fourier transform and CycleGAN as domain adaptation techniques for weed segmentation［J/OL］.Smart agricultural technology，2023，4：100188［2025-03-22］.https：//doi.org/10.1016/j.atech.2023.100188.

SANDFORT V，YAN K，PICKHARDT P J，et al.Data augmentation using generative adversarial networks （CycleGAN） to improve generalizability in CT segmentation tasks［J/OL］.Scientific reports，2019，9（1）：16884［2025-03-22］.https：//doi.org/10.1038/s41598-019-52737-x.

HU J，SHEN L，SUN G.Squeeze-and-excitation networks［C］//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. June 18-23，2018 ，Salt Lake City，UT，USA.Salt Lake City：IEEE，2018：7132-7141.

GAN X Y，FERNANDEZ I C，GUO J，et al.When to use what：methods for weighting and aggregating sustainability indicators［J］.Ecological indicators，2017，81：491-502.

代强，程曦，王永梅，等.基于轻量自动残差缩放网络的图像超分辨率重建［J］.计算机应用，2020，40（5）：1446-1452.DAI Q，CHENG X，WANG Y M，et al.Light-weight automatic residual scaling network for image super-resolution reconstruction［J］.Journal of computer applications，2020，40（5）：1446-1452.

WU Q F，CHEN Y P，MENG J.DCGAN-based data augmentation for tomato leaf disease identification［J］.IEEE access，2020，8：98716-98728.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

大黄素对重组毒株rPRRSV-DV-EGFP的抑制作用研究

生防菌根际接种调控辣椒细菌群落变化并诱导辣椒疫霉抗性

山桐子IRAP标记开发及种质的分子鉴定

牛SLC16A基因家族鉴定及SLC16A1表达特征分析

农业生物质增值化利用技术与装备团队