基于改进EfficientNet-V2的水稻病虫害识别系统研究

doi:10.3969/j.issn.1006-8082.2025.04.015

摘要/Abstract

摘要：

针对传统水稻病虫害识别方法存在的效率低下、易受主观因素干扰等局限性，以及现有深度学习模型在捕捉水稻病虫害细微特征和处理类别不平衡数据方面存在的不足，开展了一系列研究与实践。首先，利用AR眼镜在水稻田间实地采集病虫害图像，结合公开数据集IP102和网络上的图像构建了水稻病虫害数据集，并采用数据增强方式扩充训练样本，缓解类别不平衡与图像质量问题；其次，在EfficientNet-V2模型的基础上，引入CBAM注意力机制替换原有的SE模块以增强对水稻病虫害的细节特征捕捉，并采用PolyLoss损失函数优化不平衡数据学习，构建了EfficientNet-V2-Rice水稻病虫害识别模型；最后，基于改进后的识别模型，开发了一款配套的安卓手机端智能识别APP。该APP功能丰富，集成了用户注册与登录、图像上传、智能识别、识别结果检索与详情查看等核心功能模块。用户只需通过手机摄像头拍摄水稻病虫害图像并上传至APP，即可快速获得准确的识别结果，并可随时检索和查看历史识别记录的详细信息。为验证模型改进策略的有效性，进行了消融实验和对比实验。实验结果表明，本文提出的EfficientNet-V2-Rice模型在水稻病虫害识别任务中表现优秀，精确率、召回率和F₁分数分别达到84.92%、86.00%和85.45%。基于此模型开发的安卓手机端APP，为用户提供了便捷高效的识别服务，为水稻病虫害的智能监测与辅助诊断提供了一种实用的工具。

关键词: 水稻, 病虫害, AR眼镜, 图像识别, EfficientNet-V2-Rice, 手机APP

Abstract:

To address the limitations of traditional rice pest and disease recognition methods, such as low efficiency and susceptibility to subjective interference, as well as the shortcomings of existing deep learning models in capturing subtle features of rice pests and diseases and handling category-imbalanced data, a series of research and practical efforts have been undertaken. Firstly, pest and disease images were collected in rice fields using AR glasses. A rice pest and disease dataset was then constructed by combining these images with the publicly available dataset IP102 and images sourced from the web. Data augmentation techniques were employed to expand the training samples, thereby mitigating issues related to category imbalance and image quality. Secondly, based on the EfficientNet-V2 model, the CBAM (Convolutional Block Attention Module) attention mechanism was introduced to replace the original SE (Squeeze-and-Excitation) module, aiming to enhance the model’s ability to capture detailed features of rice pests and diseases. Additionally, the PolyLoss loss function was adopted to optimize the learning process for unbalanced data, leading to the construction of the EfficientNet-V2-Rice model for rice pest and disease recognition. Finally, leveraging the improved recognition model, a companion intelligent recognition APP for Android smartphones was developed. This APP boasts a rich set of features, integrating core modules such as user registration and login, image uploading, intelligent recognition, retrieval of recognition results, and viewing of detailed information. Users can simply capture images of rice pests and diseases using their smartphone cameras and upload them to the APP to quickly obtain accurate recognition results. They can also retrieve and view detailed information about historical recognition records at any time. To verify the effectiveness of the model improvement strategy, ablation and comparison experiments were conducted. The experimental results demonstrate that the proposed EfficientNet-V2-Rice model performs exceptionally well in rice pest and disease recognition tasks, achieving precision, recall, and F₁ scores of 84.92%, 86.00%, and 85.45%, respectively. The Android smartphone APP developed based on this model provides users with convenient and efficient recognition services, offering a practical tool for the intelligent monitoring and auxiliary diagnosis of rice pests and diseases.

Key words: rice, pests and diseases, AR glasses, image recognition, EfficientNet-V2-Rice, mobile APP

中图分类号:

焦佳宝, 李玲一, 刘永健, 陈相甫, 罗举, 杨保军, 姚青, 刘淑华. 基于改进EfficientNet-V2的水稻病虫害识别系统研究[J]. 中国稻米, 2025, 31(4): 86-95.

JIAO Jiabao, LI Lingyi, LIU Yongjian, CHEN Xiangfu, LUO Ju, YANG Baojun, YAO Qing, LIU Shuhua. Research on Rice Pest and Disease Recognition System Based on Improved EfficientNet-V2[J]. China Rice, 2025, 31(4): 86-95.

图/表 11

图1 利用穿戴式AR眼镜采集水稻田间病虫害图像

图2 水稻27种害虫和13种病害图像（从左到右从上到下依次为：白叶枯病Xanthomonas oryzae pv. oryzae、恶苗病Fusarium moniliforme、稻曲病Ustilaginoidea virens、稻瘟病Magnaporthe oryzae、二化螟为害状枯心Deadheart symptom caused by Chilo suppressalis、胡麻叶斑病Bipolaris oryzae、稻纵卷叶螟为害状卷叶Rolled leaf caused by Cnaphalocrocis medinalis、细菌性条斑病Xanthomonas oryzae pv. oryzicola、水稻东格鲁病毒病Rice tungro bacilliform virus、水稻细菌性褐斑病Pseudomonas oryzicola、水稻叶鞘腐败病Sarocladium oryzae、纹枯病Rhizoctonia solani、二化螟（卵）Chilo suppressalis (egg)、白背飞虱Sogatella furcifera、大稻缘蝽Leptocorisa acuta、大螟Sesamia inferens、稻赤斑沫蝉Callitetix versicolor、稻负泥甲Oulema oryzae、稻黑蝽Scotinophara lurida、稻蓟马Stenchaetothrips biformis、稻棘缘蝽Cletus punctiger、稻绿蝽Nezara viridula、稻水象甲Lissorhoptrus oryzophilus、东方粘虫Mythimna separata、稻眼蝶Mycalesis gotama、稻瘿蚊Orseoia oryzae、稻纵卷叶螟Cnaphalocrocis medinalis、电光叶蝉Inazuma dorsalis、二化螟（成虫）Chilo suppressalis (adult)、非洲蝼蛄Gryllotalpa orientalis、水稻稻苞虫（幼虫）Casinaria colacae、台湾稻螟（幼虫）Chilo auricilius (larva)、隐纹谷弄蝶Pelopidas mathias、直纹稻弄蝶Parnara guttata、中华稻蝗Oxya chinensis、福寿螺Pomacea canaliculata、灰飞虱Laodelphax striatellus、黑尾叶蝉Nephotettix cincticeps、褐飞虱Nilaparvata lugens、三化螟Scirpophaga incertulas）。

图3 数据增强结果

图4 EfficientNet-V2-Rice模型结构图

图5 MBConv-CBMA模块结构图

图6 系统组织结构图

表1 3种改进策略对EfficientNet-V2模型的影响

表2 不同模型在不同数据集对水稻病虫害识别结果（单位：%）

图7 注册、登录和图像上传功能

图8 图像识别结果列表和详情页

图9 系统管理模块

参考文献 24

[1]	赵忠权. 水稻种植与病虫害防治技术探讨[J]. 种子科技, 2025, 43(5): 169-171.
[2]	钱啸. 数据要素视角下的水稻病虫害监测技术应用集成研究[D]. 扬州: 扬州大学, 2024.
[3]	刘洋. 手机端植物病害识别与严重程度估计[D]. 兰州: 甘肃农业大学, 2021.
[4]	SHENG H Y, YAO Q, LUO J, et al. Automatic detection and counting of planthoppers on white flat plate images captured by AR glasses for planthopper field survey[J]. Computers and Electronics in Agriculture, 2024, 218: 108 639.
[5]	张志从, 崔东, 郭金锋, 等. 基于迁移学习ResNet-18的水稻病虫害识别研究[J]. 中国农学通报, 2025, 41(2): 109-116.
[6]	刘鹏, 张天翼, 冉鑫, 等. 基于PBM-YOLOv8的水稻病虫害检测[J]. 农业工程学报, 2024, 40(20): 147-156.
[7]	RAHMAN C R, ARKO P S, ALI M E, et al. Identification and recognition of rice diseases and pests using convolutional neural networks[J]. Biosystems Engineering, 2020, 194: 112-120.
[8]	NI H, SHI Z, KARUNGARU S, et al. Classification of typical pests and diseases of rice based on the ECA attention mechanism[J]. Agriculture, 2023, 13(5): 1066.
[9]	WU X P, ZHAN C, LAI Y K, et al. Ip102: A large-scale benchmark dataset for insect pest recognition[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019: 8 787-8 796.
[10]	TAN M X, LE Q V. Efficientnetv2:Smaller models and faster training[EB/OL].(2021-06-23)[2025-04-23].
[11]	HOANG V T, HOANG V D, JO K H. Rethinking mobile inverted bottleneck convolution for EfficientNet[C]//International Conference on Green Technology and Sustainable Development, 2022: 435-445.
[12]	KOONCE B. Convolutional neural networks with swift for tensorflow: Image recognition and dataset categorization[M]. Berkeley, CA, USA: Apress, 2021: 109-123.
[13]	JIE H, LI S, ALBANIE S. Squeeze-and-excitation networks[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018: 7 132-7 141.
[14]	LENG Z Q, TAN M X, LIU C X, et al. Polyloss: A polynomial expansion perspective of classification loss functions[EB/OL].(2022-05-10)[2025-04-23].
[15]	WOO S, PARK J, LEE J, et al. Cbam: Convolutional block attention module[C]//Proceedings of the European Conference on Computer Vision (ECCV), 2018: 3-19.
[16]	MAO A Q, MOHRI M, ZHONG Y T. Cross-entropy loss functions: Theoretical analysis and applications[C]//International Conference on Machine Learning, 2023: 23 803-23 828.
[17]	RUDER S. An overview of gradient descent optimization algorithms[EB/OL].(2017-06-15)[2025-04-23].
[18]	LOSHCHILOV I, HUTTER F S. Stochastic gradient descent with warm restarts. 2016 [EB/OL].(2017-05-03)[2025-04-23].
[19]	HE K, ZHANG X, REN S, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016: 770-778.
[20]	DOSOVITSKIY A, BEYER L, KOLESNIKOV A, et al. An image is worth 16×16 words:Transformers for image recognition at scale[J/OL]. arXiv.
[21]	ANDREW G, MENGLONG Z. Efficient convolutional neural networks for mobile vision applications[J]. Mobilenets, 2017, 10: 151.
[22]	ZHENG H, FU J, ZHA Z, et al. Learning deep bilinear transformation for fine-grained image representation[J]. Advances in Neural Information Processing Systems, 2019, 32: 4 279-4 288.
[23]	QIAN Y, XIAO Z, DENG Z. Fine-grained crop pest classification based on multi-scale feature fusion and mixed attention mechanisms[J]. Frontiers in Plant Science, 2025, 16: 1 500 571.
[24]	PENG Y, WANG Y. Optimizing agricultural classification with masked image modeling[J]. Cogent Food & Agriculture, 2025, 11(1): 2 462 243.