基于小样本学习的植物病害识别研究*

doi:10.13733/j.jcam.issn.20955553.2021.11.21

中国农机化学报 ›› 2021, Vol. 42 ›› Issue (11): 138-143.DOI: 10.13733/j.jcam.issn.20955553.2021.11.21

基于小样本学习的植物病害识别研究^*

肖伟¹, 冯全¹, 张建华², 杨森¹, 陈佰鸿³

1.甘肃农业大学机电工程学院,兰州市,730070;
2.中国农业科学院农业信息研究所,北京市,100081;
3.甘肃农业大学园艺学院,兰州市,730070

收稿日期:2021-04-28 修回日期:2021-09-23 出版日期:2021-11-15 发布日期:2021-11-15
通讯作者: 冯全,男,1969年生,四川隆昌人,博士,教授;研究方向为图像处理、农业信息化。E-mail: fquan@sina.com
作者简介:肖伟,男,1993年生,河南信阳人,硕士;研究方向为小样本学习的植物病害识别。E-mail: 2544760489@qq.com
基金资助:
*国家自然基金项目(31971792);甘肃省高等学校产业支撑引导项目(2019C—11)

Research on plant disease identification based on few-shot learning

Xiao Wei¹, Feng Quan¹, Zhang Jianhua², Yang Sen¹, Chen Baihong³

1. College of Mechanical and Electrical Engineering, Gansu Agricultural University, Lanzhou, 730070, China;
2. Institute of Agricultural Information, Chinese Academy of Agricultural Sciences, Beijing, 100081, China;
3. College of horticulture, Gansu Agricultural University, Lanzhou, 730070, China

Received:2021-04-28 Revised:2021-09-23 Online:2021-11-15 Published:2021-11-15

摘要/Abstract

摘要： 为在仅有少量训练样本条件下获得较高的植物病害分类精度,采用小样本学习模型作为病害分类器,在匹配网络、原型网络和关系网络3种典型小样本学习算法框架下分别采用Conv4、Conv6、ResNet10、ResNet18和ResNet34 5种浅层网络作为特征提取网络,在PlantVillage植物病害数据集上对病害识别性能进行对比试验。在1shot条件下,匹配网络、原型网络和关系网络对植物叶片病害识别的平均准确率分别为72.29%、72.43%和69.45%;其中原型网络+ResNet34为表现最好的组合,病害识别准确率达到了77.60%。在5shot条件下,匹配网络、原型网络和关系网络平均准确率分别为87.11%、87.50%和82.92%,各种网络病害识别准确率比1shot条件均有明显提升;原型网络+ResNet34依旧是表现最佳的组合方式,识别准确率达到89.66%。上述试验结果表明,通过优选小样本学习框架和特征提取网络的组合方式,对于少量样本的病害也能取得较好的识别效果。

关键词: 小样本学习, 特征提取网络, 分类器, 植物病害, 识别效果

Abstract: In order to obtain high accuracy of plant disease classification with only a few training samples, a few-shot learning model was used as the disease classifier, and five kinds of shallow networks, Conv4, Conv6, ResNet10, ResNet18, and ResNet34, were used as feature extraction networks under the framework of three typical few-shot learning algorithms, including MatchingNet, ProtoNet, and RelationNet. Their performances were compared on the plant disease data set of PlantVillage. Under the condition of 5way、1shot, the average accuracies of MatchingNet, ProtoNet, and RelationNet were 72.29%, 72.43%, and 69.45%, respectively. ProtoNet+ResNet34 was the optimal combinational mode, and the accuracy reached 77.60%. Under the condition of 5way、5shot, the average accuracies of MatchingNet, ProtoNet, and RelationNet were 87.11%, 87.50%, and 82.92%, respectively. The accuracies were significantly improved compared to that of the 1shot condition. ProtoNet+ResNet34 was still the optimal one with an accuracy of 89.66%. The above test results show that by optimizing the combination of a few-shot learning framework and feature extraction network, the recognition model can achieve good effects for diseases with a small number of samples.

Key words: few-shot learning, feature extraction network, classifier, plant disease, identification effect

中图分类号:

肖伟, 冯全, 张建华, 杨森, 陈佰鸿. 基于小样本学习的植物病害识别研究^*[J]. 中国农机化学报, 2021, 42(11): 138-143.

Xiao Wei, Feng Quan, Zhang Jianhua, Yang Sen, Chen Baihong. Research on plant disease identification based on few-shot learning[J]. Journal of Chinese Agricultural Mechanization, 2021, 42(11): 138-143.

参考文献

[1] 魏丽冉, 岳峻, 李振波, 等. 基于核函数支持向量机的植物叶部病害多分类检测方法[J]. 农业机械学报, 2017, 48(S1): 166-171.
Wei Liran, Yue Jun, Li Zhenbo, et al. Multi-classification detection method for plant leaf diseases based on kernel function SVM [J]. Transactions of the Chinese Society of Agricultural Machinery, 2017, 48(S1): 166-171.
[2] 马浚诚, 杜克明, 郑飞翔, 等. 基于卷积神经网络的温室黄瓜病害识别系统[J]. 农业工程学报, 2018, 34(12): 186-192.
Ma Juncheng, Du Keming, Zheng Feixiang, et al.Disease recognition system for greenhouse cucumbers based on deep convolutional neural network [J]. Transactions of the Chinese Society of Agricultural Engineering, 2018, 34(12): 186-192.
[3] 孙俊, 谭文军, 毛罕平. 基于改进卷积神经网络的多种植物叶片病害识别[J]. 农业工程学报, 2017, 33(19): 209-215.
Sun Jun, Tan Wenjun, Mao Hanping.Recognition of multiple plant leaf diseases based on improved convolutional neural network [J]. Transactions of the Chinese Society of Agricultural Engineering, 2017, 33(19): 209-215.
[4] Mohanty S P, Hughes D P, Salathe M. Using deep learning for image-based plant disease detection [J]. Frontiers in plant Science, 2016, 7: 1-10.
[5] Amara J, Bouaziz B, Algergawy A, et al. A deep learning-based approach for banana leaf diseases classification [C]. Gesellschaft Fur Informatik, Bonn: Lecture Notes in Informatics, 2017: 79-88.
[6] Brahimi M, Boukhalfa K, Moussaoui A. Deep learning for tomato disease: Classification and symptoms visualization [J]. Applied Artificial Intelligence, 2017, 31(4): 299-315.
[7] Liu B, Zhang Y, He D, et al. Identification of apple leaf diseases based on deep convolutional neural networks [J]. Symmetry, 2018, 10(1): 1-16.
[8] Dyrmann M, Karstoft H, Midtiby H S. Plant species classification using deep convolutional neural network [J]. Biosystems Engineering, 2016, 151: 72-80.
[9] 张建华, 孔繁涛, 吴建寨, 等. 基于改进VGG卷积神经网络的棉花病害识别模型[J]. 中国农业大学学报, 2018, 23(11): 161-171.
Zhang Jianhua, Kong Fantao, Wu Jianzhai, et al.Cotton disease identification model based on improved VGG convolution neural network [J]. Journal of China Agricultural University, 2018, 23(11): 161-171.
[10] 郑一力, 张露. 基于迁移学习的卷积神经网络植物叶片图像识别方法[J]. 农业机械学报, 2018, 49(S1): 354-359.
Zheng Yili, Zhang Lu. Plant leaf image recognition method based on transfer learning with convolutional neural networks [J]. Transactions of the Chinese Society for Agricultural Machinery, 2018, 49(S1): 354-359.
[11] Vinyals O, Blundell C, Lillicrap T, et al. Matching network for one shot learning [J]. In Advances in Neural Information Processing System, 2016, 7: 3630-3638.
[12] Sung F, Yang Y, Zhang L. Learning to compare: Relation network for few-shot learning [C]. Hawaii USA: IEEE Conference on Computer Vision and Pattern Recognition, 2017: 1199-1208.
[13] 任胜男, 孙钰, 张海燕, 等. 基于one-shot学习的小样本植物病害识别[J]. 江苏农业学报, 2019, 35(5): 1061-1067.
Ren Shengnan, Sun Yu, Zhang Haiyan, et al.Plant disease identification for small sample based on one-shot learning [J]. Jiangsu Journal of Agriculture Sciences, 2019, 35(5): 1061-1067.
[14] Snell J, Swersky K, Zemel R S. Prototypical networks for few-shot learning [C]. Long Beach: Thirty-first Conference in Neural Information Processing Systems, 2017: 4077-4087.
[15] He K M, Zhang X G, Ren S Q, et al. Deep residual learning for image recognition [C]. IEEE conference on computer vision and pattern recognition, IEEE computer society, 2016: 770-778.

基于小样本学习的植物病害识别研究^*

Research on plant disease identification based on few-shot learning

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 4

编辑推荐

Metrics

本文评价

[1]	项小东;翟蔚;黄言态;刘薇;. 基于Xception-CEMs神经网络的植物病害识别[J]. 中国农机化学报, 2021, 42(8): 177-186.
[2]	李健;熊琦;胡雅婷;. 基于改进AdaBoost算法对环柄菇毒性判别研究[J]. 中国农机化学报, 2021, 42(3): 72-77.
[3]	蔡浩;郭宏亮;. 基于多分类器DS证据理论融合的水果识别研究[J]. 中国农机化学报, 2021, 42(2): 184-189.
[4]	樊湘鹏;周建平;许燕;杨晶晶;. 数据集对基于深度学习的作物病害识别有效性影响[J]. 中国农机化学报, 2021, 42(1): 192-200.