|本期目录/Table of Contents|

[1]孔江波,朱贺,张延文,等.基于实例置信度推断的半监督小样本植物病害图像识别[J].江苏农业科学,2022,50(12):229-237.
　Kong Jiangbo,et al.Semi-supervised few-shot plant disease image recognition based on instance credibility inference[J].Jiangsu Agricultural Sciences,2022,50(12):229-237.
点击复制

基于实例置信度推断的半监督小样本植物病害图像识别(PDF)
分享到：

《江苏农业科学》[ISSN:1002-1302/CN:32-1214/S]

卷:
第50卷
期数:
2022年第12期
页码:
229-237
栏目:
农业工程与信息技术
出版日期:
2022-06-20

文章信息/Info

Title:
Semi-supervised few-shot plant disease image recognition based on instance credibility inference
作者:
孔江波; 朱贺; 张延文; 黄建平
东北林业大学机电工程学院，黑龙江哈尔滨 150040
Author(s):
Kong Jiangbo; et al
关键词:
植物病害; 半监督学习; 识别; 小样本学习; 深度学习; 置信度推断
Keywords:
-
分类号:
TP391.41；S126
DOI:
-
文献标志码:
A
摘要:
植物病害影响农业生产的产量和质量。针对现有小样本植物病害识别方法大多数都是基于监督式学习模型以及少数半监督学习方法未判别伪标注样本的可信度的问题，提出一种基于实例置信度推断的半监督学习小样本植物病害图像识别方法。首先构建ResNet-12网络提取有标注样本和无标注样本的特征；其次利用极少数有标签的植物病害样本训练SVM分类器，用分类器推断无标注样本的类别并赋予伪标签；然后采用实例置信度推断(ICI)算法获取伪标注样本的置信度，迭代选择可信的伪标签样本加入训练拓展支持集；最后应用训练后的网络模型对植物病害图像进行识别。该方法能够反映无标注病害样本的真实分布，迭代选择最可信的伪标注样本进行模型训练，从而提高模型的识别性能。试验采用Plant Village公开数据集进行10-way-5-shot试验。结果显示：在unlabel= 50的情况下识别准确率为89.34%，病害的各项评价指标均随着无标注样本数量的增加而增加。结果表明本研究提出的方法从无标注样本中获取到的信息是鲁棒的，且识别准确率优于传统迁移学习，能有效提升小样本条件下植物病害图像的识别效果。
Abstract:
-

参考文献/References:

［1］Kaur S，Pandey S，Goel S. Plants disease identification and classification through leaf images：a survey［J］. Archives of Computational Methods in Engineering，2019，26(2)：507-530.
［2］翟肇裕，曹益飞，徐焕良，等. 农作物病虫害识别关键技术研究综述［J］. 农业机械学报，2021，52(7)：1-18.
［3］Strange R N，Scott P R. Plant disease：a threat to global food security［J］. Annual Review of Phytopathology，2005，43：83-116.
［4］Iqbal Z，Khan M A，Sharif M，et al. An automated detection and classification of citrus plant diseases using image processing techniques：a review［J］. Computers and Electronics in Agriculture，2018，153：12-32.
［5］Mahlein A K. Plant disease detection by imaging sensors-parallels and specific demands for precision agriculture and plant phenotyping［J］. Plant Disease，2016，100(2)：241-251.
［6］Liu J，Wang X W. Plant diseases and pests detection based on deep learning：a review［J］. Plant Methods，2021，17(1)：22.
［7］孙红，李松，李民赞，等. 农业信息成像感知与深度学习应用研究进展［J］. 农业机械学报，2020，51(5)：1-17.
［8］Abade A，Ferreira P A，de Barros V F. Plant diseases recognition on images using convolutional neural networks：a systematic review［J］. Computers and Electronics in Agriculture，2021，185：106125.
［9］Lu J Z，Tan L J，Jiang H Y. Review on convolutional neural network (CNN) applied to plant leaf disease classification［J］. Agriculture，2021，11(8)：707.
［10］Sladojevic S，Arsenovic M，Anderla A，et al. Deep neural networks based recognition of plant diseases by leaf image classification［J］. Computational Intelligence and Neuroscience，2016，2016：3289801.
［11］黄建平，陈镜旭，李克新，等. 基于神经结构搜索的多种植物叶片病害识别［J］. 农业工程学报，2020，36(16)：166-173.
［12］Argüeso D，Picon A，Irusta U，et al. Few-shot learning approach for plant disease classification using images taken in the field［J］. Computers and Electronics in Agriculture，2020，175：105542.
［13］Li Y，Chao X W. Semi-supervised few-shot learning approach for plant diseases recognition［J］. Plant Methods，2021，17(1)：68.
［14］Vinyals O，Blundell C，Lillicrap T P，et al. Matching networks for one shot learning［C］//NeurIPS16：Proceedings of the 30th International Conference on Neural Information Processing Systems. Barcelona，Spain：ACM，2016：3630-3638.
［15］Snell J，Swersky K，Zemel R S. Prototypical networks for few-shot learning［C］//NeurIPS17：Proceedings of the 31st International Conference on Neural Information Processing Systems(NeurIPS). Long Beach：ACM，2017：4077-4087.
［16］刘颖，雷研博，范九伦，等. 基于小样本学习的图像分类技术综述［J］. 自动化学报，2021，47(2)：297-315.
［17］赵凯琳，靳小龙，王元卓. 小样本学习研究综述［J］. 软件学报，2021，32(2)：349-369.
［18］Zhu X J，Goldberg A B. Introduction to semi-supervised learning［J］. Synthesis Lectures on Artificial Intelligence and Machine Learning，2009，3(1)：1-130.
［19］Goodfellow I，Pouget-Abadie J，Mirza M，et al. Generative adversarial networks［J］. Communications of the ACM，2020，63(11)：139-144.
［20］Pan S J，Yang Q. A survey on transfer learning［J］. IEEE Transactions on Knowledge and Data Engineering，2010，22(10)：1345-1359.
［21］Hariharan B，Girshick R. Low-shot visual recognition by shrinking and hallucinating features［C］//2017 IEEE International Conference on Computer Vision(ICCV).Venice：IEEE，2017：3037-3046.
［22］Qi H，Brown M，Lowe D G. Low-shot learning with imprinted weights［C］//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Salt Lake City：IEEE，2018：5822-5830.
［23］Finn C，Abbeel P，Levine S. Model-agnostic meta-learning for fast adaptation of deep networks［C］//ICML17：Proceedings of the 34th International Conference on Machine Learning-Volume 70.2017：1126-1135.
［24］Li Z G，Zhou F W，Chen F，et al. Meta-SGD：learning to learn quickly for few-shot learning［EB/OL］. ［2021-04-07］. https：//arxiv.org/abs/1707.09835.
［25］Nichol A，Achiam J，Schulman J. On first-order meta-learning algorithms［EB/OL］. ［2021-04-07］. https：//arxiv.org/abs/1803.02999.
［26］Wang B，Wang D. Plant leaves classification：a few-shot learning method based on Siamese network［J］. IEEE Access，2019，7：151754-151763.
［27］Yu H J，Son C H. Leaf spot attention network for apple leaf disease identification［C］//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).Seattle：IEEE，2020：229-237.
［28］Hu G S，Wu H Y，Zhang Y，et al. A low shot learning method for tea leafs disease identification［J］. Computers and Electronics in Agriculture，2019，163：104852.
［29］谢军，江朝晖，李博，等. 基于二次迁移模型的小样本茶树病害识别［J］. 江苏农业科学，2021，49(6)：176-182.
［30］周志华. 机器学习 ［M］. 北京：清华大学出版社，2016.
［31］Wang Y K，Xu C M，Liu C，et al. Instance credibility inference for few-shot learning［C］//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Seattle：IEEE，2020：12833-12842.
［32］Fan J Q，Tang R L，Shi X F.Partial consistency with sparse incidental parameters［J］. Statistica Sinica，2018，28：2633-2655.
［33］Simon N，Friedman J，Hastie T. A blockwise descent algorithm for group-penalized multiresponse and multinomial regression［EB/OL］. ［2021-04-07］. https：//arxiv.org/abs/1311.6529.
［34］Oreshkin B N，Rodriguez P，Lacoste A. Tadam：Task dependent adaptive metric for improved few-shot learning［C］//NeurIPS 18：Proceedings of the 32st International Conference on Neural Information Processing Systems. Montréal：ACM，2018：719-729.
［35］Mishra N，Rohaninejad M，Chen X，et al. A simple neural attentive meta-learner［C］//ICLR 2018. Vancouver，2018.
［36］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.Las Vegas：IEEE，2016：770-778.
［37］Joachims T. Transductive inference for text classification using support vector machines［C］//ICML 1999：Proceedings of the 16th International Conference on Machine Learning. Bled，1999：200-209.
［38］Vapnik V. Transductive inference and semi-supervised learning ［J］. Semi-Supervised Learning. Cambridge：MIT Press，2006：453-472.

相似文献/References:

[1]马国胜,陈娟,薛毅.苏南地区绿篱病害发生规律与生态控制技术[J].江苏农业科学,2013,41(11):131.
　Ma Guosheng,et al.Occurrence and eco-control technology of hedgerow diseases in Southern Jiangsu[J].Jiangsu Agricultural Sciences,2013,41(12):131.
[2]易龙,张亚,廖晓兰,等.链霉菌防治植物病害的研究进展[J].江苏农业科学,2014,42(03):91.
　Yi Long,et al.Research progress of streptomyces controlling plant diseases[J].Jiangsu Agricultural Sciences,2014,42(12):91.
[3]王超,郭坚华,席运官,等.拮抗细菌在植物病害生物防治中应用的研究进展[J].江苏农业科学,2017,45(18):1.
　Wang Chao,et al.Research progress on application of antagonistic bacteria in biological control of plant diseases[J].Jiangsu Agricultural Sciences,2017,45(12):1.
[4]宋薇薇,朱辉,余凤玉,等.植物内生菌及其对植物病害的防治作用综述[J].江苏农业科学,2018,46(06):12.
　Song Weiwei,et al.Plant endophytes and their control effects on plant diseases：a review[J].Jiangsu Agricultural Sciences,2018,46(12):12.
[5]何宇,吕卫光,张娟琴,等.生防菌对稻瘟病害控制的研究进展[J].江苏农业科学,2021,49(21):40.
　He Yu,et al.Research progress of biocontrol bacteria controlling rice blast disease[J].Jiangsu Agricultural Sciences,2021,49(12):40.
[6]杨祥,段军明,董明刚.面向移动端的植物病害图像识别方法及其应用[J].江苏农业科学,2023,51(4):191.
　Yang Xiang,et al.Mobile-oriented plant disease image recognition method and its application[J].Jiangsu Agricultural Sciences,2023,51(12):191.
[7]代国威,胡林,樊景超,等.基于GLCM特征提取和投票分类模型的马铃薯早、晚疫病检测[J].江苏农业科学,2023,51(8):185.
　Dai Guowei,et al.Detection of potato early and late blight based on GLCM feature extraction and voting classification model[J].Jiangsu Agricultural Sciences,2023,51(12):185.
[8]郑旭康,李志忠,秦俊豪.基于半监督学习的梨叶病害检测[J].江苏农业科学,2024,52(5):192.
　Zheng Xukang,et al.Study on pear leaf disease detection based on semi-supervised learning[J].Jiangsu Agricultural Sciences,2024,52(12):192.
[9]严露露,朱赞彬,冯世杰,等.基于改进FixMatch算法的半监督番茄病虫害识别[J].江苏农业科学,2024,52(20):244.
　Yan Lulu,et al.Semi-supervised identification of tomato diseases and pests based on improved FixMatch algorithm[J].Jiangsu Agricultural Sciences,2024,52(12):244.

备注/Memo

备注/Memo:
收稿日期：2021-11-16
基金项目：中央高校基本科研业务费专项(编号：2572019CP19)；黑龙江省自然科学基金(编号：TD2020C001)。
作者简介：孔江波(1997—)，男，甘肃永靖人，硕士研究生，主要从事模式识别研究。E-mail：kjb1206612461@163.com。
通信作者：黄建平，博士，副教授，博士生导师，主要从事目标检测、图像处理相关研究。E-mail：jphuang@nefu.edu.cn。
更新日期/Last Update: 2022-06-20