[1]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,(05):1061-1067.[doi:doi:10.3969/j.issn.1000-4440.2019.05.009]
　REN Sheng-nan,SUN Yu,ZHANG Hai-yan,et al.Plant disease identification for small sample based on one-shot learning[J].,2019,(05):1061-1067.[doi:doi:10.3969/j.issn.1000-4440.2019.05.009]
点击复制

基于one-shot学习的小样本植物病害识别()
分享到：

江苏农业学报[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2019年05期
页码:
1061-1067
栏目:
植物保护
出版日期:
2019-10-31

文章信息/Info

Title:
Plant disease identification for small sample based on one-shot learning
作者:
任胜男1; 孙钰1; 张海燕1; 郭丽霞2
(1.北京林业大学信息学院，北京100083; 2.国家食品安全风险评估中心，北京100022)
Author(s):
REN Sheng-nan1; SUN Yu1; ZHANG Hai-yan1; GUO Li-xia2
(1.School of Information Science and Technology, Beijing Forestry University, Beijing 100083, China; 2.China National Center for Food Safety Risk Assessment, Beijing 100022, China)
关键词:
植物病害识别; 深度学习; one-shot学习; 焦点损失函数; 关系网络
Keywords:
plant disease identification;  deep learning;  one-shot learning;  focal loss;  relation network
分类号:
S126
DOI:
doi:10.3969/j.issn.1000-4440.2019.05.009
文献标志码:
A
摘要:
针对植物病害小样本问题提出一种基于one-shot学习的植物病害识别方法。以公开数据集PlantVillage中8类样本数量较少的植物病害图像作为识别对象，使用焦点损失函数(focal loss, FL)训练基于关系网络的植物病害分类器。训练过程中，调整FL超参数使模型聚焦于困难样本，从而提高植物病害识别精确率。结果表明：该方法在5-way、1-shot任务中识别精确率达到89.90%，相比原始关系网络模型精确率提高了4.69个百分点。同时，与匹配网络和迁移学习相比，改进后的方法在实验数据集上识别精确率分别提高了25.02个百分点和41.90个百分点。
Abstract:
A plant disease identification method based on one-shot learning was proposed to solve the problem of small plant disease dataset. Using eight kinds of plant disease images with few samples in the public dataset PlantVillage as recognition objects, the disease classifier based on relation network was trained by using focal loss (FL). To improve the accuracy of plant disease identification, the FL hyper-parameters were adjusted to focus the model on the hard samples during the training process. The results showed that the recognition accuracy of this method was 89.90% in 5-way and 1-shot task, which was 4.69% higher than that of the original relation network model. Besides, compared with matching network and transfer learning, the accuracy of improved method was improved by 25.02% and 41.90% on experimental set.

参考文献/References:

[1]王华弟， 沈颖， 黄茜斌， 等. 浙江省杨梅病虫害种类与发生规律及其绿色防控技术[J]. 南方农业学报 ，2017，48(4)：640-646.
[2]王震，李金秀，张彬，等.赤霉病对江苏省北部小麦主栽品种品质的影响[J]．江苏农业科学，2018，46(13)：105-108.
[3]周云成，许童羽，郑伟，等.基于深度卷积神经网络的番茄主要器官分类识别方法[J].农业工程学报,2017,33(15):219-226.
[4]王璨，武新慧，李志伟，等.基于卷积神经网络提取多尺度分层特征识别玉米杂草[J].农业工程学报,2018,34(5):144-151.
[5]刘阗宇，冯全.基于卷积神经网络的葡萄叶片检测[J].西北大学学报(自然科学报)，2017，47(4):505-512.
[6]孙俊，谭文军，毛罕平，等.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.
[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]AMARA J, BOUAZIZ B, ALGERGAWY A, et al. A deep learning-based approach for banana leaf diseases classification[C].Gesellschaft Für Informatik,Bonn:Lecture Notes in Informatics,2017:79-88.
[9]BRAHIMI M, BOUKHALFA K, MOUSSAOUI A. Deep learning for tomato disease:Classification and symptoms visualization[J].Applied Artificial Intelligence,2017,31(4):299-315.
[10]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.
[11]CRUZ A, LUVISI A, BELLIS L D, et al. X-FIDO:An effective application for detecting olive quick decline syndrome with deep learning and data fusion[J].Frontiers in Plant Science,2017,8:1741.
[12]FERENTINOS K P. Deep learning models for plant disease detection and diagnosis[J].Computers and Electronics in Agriculture,2018,145:311-318.
[13]LI F F, FERGUS R, PERONA P. One-shot learning of object categories[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2006,28(4):594-611.
[14]LAKE B M, SALAKHUTDINOV R, TENENBAUM J B. Human-level concept learning through probabilistic program induction[J].Science,2015,350(6266):1332-1338.
[15]FINN C, ABBEEL P, LEVINE S.Model-agnostic meta-Learning for fast adaptation of deep networks[C].Sydney Australia:International Conference on Machine Learning,2017:1126-1135.
[16]RAVI S, LAROCHELLE H. Optimization as a model for few-shot learning[C].San Juan,Puerto Rico:International Conference on Learning Representation,2016:1-11.
[17]KOCH G, ZEMEL R, SALAKHUTDINOV R. Siamese neural networks for One-shot image recognition[C].Lille France:International Conference on Machine Learning,2015:1-8.
[18]VINYALS O, BLUNDELL C, LILLICRAP T, et al. Matching network for one shot learning[J].In Advances in Neural Information Processing Systems,2016,7:3630-3638.
[19]SNELL J, SWERSKY K, ZEMEL R S. Prototypical networks for few-shot learning[C].Long Beach:Thirty-first Conference on Neural Information Processing Systems,2017:4077-4087.
[20]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.
[21]LIN T Y, GOYAL P, GIRSHICK R, et al. Focal loss for dense object detection[C].Venice Italy:International Conference on Computer Vision,2017:2999-3007.

相似文献/References:

[1]宋晓倩,张学艺,张春梅,等.基于深度迁移学习的酿酒葡萄种植信息提取[J].江苏农业学报,2020,(03):689.[doi:doi:10.3969/j.issn.1000-4440.2020.03.022]
　SONG Xiao-qian,ZHANG Xue-yi,ZHANG Chun-mei,et al.Extraction of wine grape planting information based on deep transfer learning[J].,2020,(05):689.[doi:doi:10.3969/j.issn.1000-4440.2020.03.022]
[2]金寿祥,周宏平,姜洪喆,等.采摘机器人视觉系统研究进展[J].江苏农业学报,2023,(02):582.[doi:doi:10.3969/j.issn.1000-4440.2023.02.033]
　JIN Shou-xiang,ZHOU Hong-ping,JIANG Hong-zhe,et al.Research progress on visual system of picking robot[J].,2023,(05):582.[doi:doi:10.3969/j.issn.1000-4440.2023.02.033]
[3]李颀,郭梦媛.基于深度学习的休眠期苹果树点云语义分割[J].江苏农业学报,2023,(05):1189.[doi:doi:10.3969/j.issn.1000-4440.2023.05.011]
　LI Qi,GUO Meng-yuan.Semantic segmentation of apple tree point cloud in dormant period based on deep learning[J].,2023,(05):1189.[doi:doi:10.3969/j.issn.1000-4440.2023.05.011]
[4]翟先一,魏鸿磊,韩美奇,等.基于改进YOLO卷积神经网络的水下海参检测[J].江苏农业学报,2023,(07):1543.[doi:doi:10.3969/j.issn.1000-4440.2023.07.011]
　ZHAI Xian-yi,WEI Hong-lei,HAN Mei-qi,et al.Underwater sea cucumber identification based on improved YOLO convolutional neural network[J].,2023,(05):1543.[doi:doi:10.3969/j.issn.1000-4440.2023.07.011]
[5]许鑫,耿庆,郑凯,等.基于纹理特征与深度学习的小麦图像中的穗粒分割与计数[J].江苏农业学报,2024,(04):661.[doi:doi:10.3969/j.issn.1000-4440.2024.04.010]
　XU Xin,GENG Qing,ZHENG Kai,et al.Segmentation and counting of wheat spikes and grains based on texture features and deep learning[J].,2024,(05):661.[doi:doi:10.3969/j.issn.1000-4440.2024.04.010]
[6]李仁杰,宋涛,高婕,等.基于改进YOLOv5的自然环境下番茄患病叶片检测模型[J].江苏农业学报,2024,(06):1028.[doi:doi:10.3969/j.issn.1000-4440.2024.06.009]
　LI Renjie,SONG Tao,GAO Jie,et al.Tomato diseased leaf detection model based on improved YOLOv5 in natural environment[J].,2024,(05):1028.[doi:doi:10.3969/j.issn.1000-4440.2024.06.009]

备注/Memo

备注/Memo:
收稿日期：2018-12-03 基金项目：北京林业大学中央高校基本科研业务费专项资金资助项目(TD2014-01)；国家重点研发计划项目(2017YFC1602002) 作者简介：任胜男(1993-)，女，山东济南人，硕士，研究方向物联网技术。(E-mail)18810977315@163.com 通讯作者：张海燕，(E-mail)zhyzml@bjfu.edu.cn
更新日期/Last Update: 2019-11-11