基于ResNet-50深度卷积网络的果树病害智能诊断模型研究

doi:10.13998/j.cnki.issn1002-1248.20-0670

农业图书情报学报 ›› 2021, Vol. 33 ›› Issue (4): 58-67.doi: 10.13998/j.cnki.issn1002-1248.20-0670

所属专题：农业知识服务

基于ResNet-50深度卷积网络的果树病害智能诊断模型研究

金瑛¹, 叶飒^2,*, 李洪磊¹

1.辽宁师范大学大数据与商务智能实验室,大连 116029;
2.中国农业科学院农业信息研究所,北京 100081

收稿日期:2020-07-28 出版日期:2021-04-05 发布日期:2021-04-30
通讯作者: ^*叶飒(ORCID:0000-0002-1339-9060),女,馆员,中国农业科学院农业信息研究所,研究方向为数值分析与数据挖掘。Email:yesa@caas.cn
作者简介:金瑛(ORCID:0000-0001-7973-3693),女,硕士研究生,辽宁师范大学管理科学与工程,研究方向为系统建模与优化。李洪磊(ORCID:0000-0002-7080-873X),男,博士,特聘教授,硕士研究生导师,辽宁师范大学,研究方向为系统建模与仿真、智能信息处理
基金资助:
中国工程科技知识中心建设项目“农业专业知识服务系统”(CKCEST-2020-1-20)

The Intelligent Diagnosis Model of Fruit Tree Disease Based on ResNet-50

JIN Ying¹, YE Sa^2,*, LI Honglei¹

1. School of Government Management, Liaoning Normal University, Dalian 116029;
2. Institute of Agricultural Information, Chinese Academy of Agricultural Sciences, Beijing 100081

Received:2020-07-28 Online:2021-04-05 Published:2021-04-30

摘要/Abstract

摘要： [目的/意义]果树病害危及农业生产安全,运用人工智能技术帮助果农及时准确地识别果树病害对保障农业安全生产具有重要意义。[方法/过程]采用10 000张果树叶片病斑图像数据集,通过旋转、污化、增噪、切割等图像增强手段,提高样本图像的多样性;使用ResNet-50深度卷积网络模型,进行机器学习,获得果树病害识别模型,并基于此模型开发了应用软件提供在线诊断服务。[结果/结论]实验结果表明:该模型对4种果树病害的平均识别率达到92.9%,和相关研究成果相比具有较好的诊断效果。

关键词: ResNet-50, 图像识别, 果树疾病, 智能诊断

Abstract: [Purpose/Significance] Fruit tree diseases endanger the safety of agricultural production, and the use of artificial intelligence technologies to help fruit growers identify fruit tree diseases in a timely and accurate manner is of great significance to ensure safe agricultural production. [Method/Process] Using 10 000 fruit tree leaf diseased spots image data sets, through image enhancement methods such as rotation, pollution, noise enhancement, and cutting to improve the diversity of sample images; using the ResNet-50 deep convolutional network model to perform machine learning to obtain the fruit tree diseases identification model, and develop application software based on this model to provide online diagnostic services. [Results/Conclusions] The experimental results show that the average recognition rate of the four fruit tree diseases reached 92.9%, which has a better diagnostic effect compared with related research results.

Key words: ResNet-50, image recognition, fruit tree disease, intelligent diagnosis

中图分类号:

G249.2

金瑛, 叶飒, 李洪磊. 基于ResNet-50深度卷积网络的果树病害智能诊断模型研究[J]. 农业图书情报学报, 2021, 33(4): 58-67.

JIN Ying, YE Sa, LI Honglei. The Intelligent Diagnosis Model of Fruit Tree Disease Based on ResNet-50[J]. Journal of Library and Information Science in Agriculture, 2021, 33(4): 58-67.

参考文献

[1] 祁钊, 江朝晖, 杨春合, 等. 基于图像技术的玉米叶部病害识别研究[J]. 安徽农业大学学报, 2016, 43(2):167-172.
QI Z, JIANG Z H, YANG C H, et al.Identification of maize leaf diseases based on image technology[J]. Journal of Anhui agricultural university, 2016, 43(2): 167-172.
[2] 刘浩洲, 陈礼鹏, 穆龙涛, 等. 基于K-means聚类的猕猴桃花朵识别方法[J]. 农机化研究, 2020, 42(2): 22-26.
LIU H Z, CHEN L P, MU L T, et al.A recognition method of kiwifruit flowers based on K-means clustering[J]. Journal of agricultural mechanization research, 2020, 42(2): 22-26.
[3] 毕傲睿. 苹果叶子病害图像识别系统的设计与实现[D]. 西安: 西安建筑科技大学, 2014.
BI A R.Design and implementation of apple-leaf disease image recognition system[D]. Xi'an: Xi'an university of architecture and 4 technology, 2014.
[4] 屈赟, 陶晡, 王政嘉, 等. 基于Android的苹果叶部病害识别系统设计[J]. 河北农业大学学报, 2015, 38(6): 102-106.
QU Y, TAO B, WANG Z J, et al.Design of apple leaf disease recognition system based on Android[J]. Journal of Hebei agricultural university, 2015, 38(6): 102-106.
[5] 刘双. 基于数字图像的农业害虫精准检测算法的研究——以菜蝽检测为例[D]. 雅安: 四川农业大学, 2014.
LIU S.Study on exact pest detection algorithm based on digital image - Illustrated by the case of Eurydema dominulus. Yaan: Sichuan agricultural university, 2014.
[6] 卢柳江, 匡迎春, 陈兰鑫, 等. 基于级联AdaBoost分类器的农作物虫害图像识别研究[J]. 中国农机化学报, 2019, 40(8): 127-131.
LU L J, KUANG Y C, CHEN L X, et al.Research on pest recogni-tion based on cascaded AdaBoost classifier[J]. Journal of Chinese agricultural mechanization, 2019, 40(8): 127-131.
[7] LECUN Y, BENGIO Y, HINTON G.Deep learning[J]. Nature, 2015, 521(7553): 436-444.
[8] 吕盛坪, 李灯辉, 冼荣亨. 深度学习在我国农业中的应用研究现状[J]. 计算机工程与应用, 2019, 55(20): 24-33, 51.
LV S P, LI D H, XIAN R H.Research Status of Deep Learning in Agriculture of China[J]. Computer engineering and applications, 2019, 55(20): 24-33, 51.
[9] 黄双萍, 孙超, 齐龙, 等. 基于深度卷积神经网络的水稻穗瘟病检测方法[J]. 农业工程学报, 2017, 33(20):169-176.
HUANG S P, SUN C, QI L, et al.Rice panicle blast identification method based on deep convolution neural network[J]. Transactions of the Chinese society of agricultural engineering, 2017, 33(20): 169-176.
[10] 李艳. 基于改进CNN的马铃薯病害识别算法[J]. 信息通信, 2017, 6: 46-48.
LI Y.Potato disease recognition algorithm based on improved CNN[J]. Information & communications, 2017, 6: 46-48.
[11] 李淼, 王敬贤, 李华龙, 等. 基于CNN和迁移学习的农作物病害识别方法研究[J]. 智慧农业, 2019, 1(3): 46-55.
LI M, WANG J X, LI H L, et al.Method for identifying crop disease based on CNN and transfer learning[J]. Smart agriculture, 2019, 1(3): 46-55.
[12] 王细萍, 黄婷, 谭文学, 等. 基于卷积网络的苹果病变图像识别方法[J]. 计算机工程, 2015, 41(12): 293-298.
WANG X P, HUANG T, TAN W X, et al.Apple lesion image recognition method based on convolutional network[J]. Computer engineering, 2015, 41(12): 293-298.
[13] 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): 167-177.
ZHANG J H, KONG F T, WU J Z, et al.Cotton disease identification model based on improved VGG convolution neural network[J]. Journal of China agricultural university, 2018, 23(11): 167-177.
[14] 蒋丰千, 李旸, 余大为, 等. 基于Caffe卷积神经网络的大豆病害检测系统[J]. 浙江农业学报, 2019, 31(7): 1177-1183.
JIANG F Q, LI Y, YU D W, et al.Soybean disease detection system based on convolutional neural network under Caffe framework[J]. Acta agriculturae Zhejiangensis, 2019, 31(7): 1177-1183.
[15] 王梅嘉, 何东健, 任嘉琛. 基于Android平台的苹果叶病害远程识别系统[J]. 计算机工程与设计, 2015, 36(9): 2585-2590.
WANG M J, HE D J, REN J C.Remote recognition of apple leaf disease based on Android platform[J]. Computer engineering and design, 2015, 36(9): 2585-2590.
[16] 邱靖, 刘继荣, 曹志勇, 等. 基于卷积神经网络的水稻病害图像识别研究[J]. 云南农业大学学报(自然科学), 2019, 34(5): 884-888.
QIU J, LIU J R, CAO Z Y, et al.Rice disease image recognition research based on convolutional neural network[J]. Journal of Yun-nan agricultural university (natural science), 2019, 34(5): 884-888.
[17] 中国工程科技知识中心. 农业专业知识服务系统作物病虫害图谱库[EB/OL].[2020-05-21]. http://agri.ckcest.cn/specialtyresources/list29-1.html.
CKCEST. Image database of crop disease of agricultural knowledge service system[EB/OL]. [2020-05-21]. http://agri.ckcest.cn/specialtyresources/list29-1.html.
[18] BRUZZONE L, SERPICO S B.Classification of imbalanced remote-sensing data by neural networks[J]. Pattern recognition letters, 1997, 18(11): 323-1328.
[19] FRANCISCO J P, ANTONIO J R, FRANCISCO C, et al.On the impact of imbalanced data in convolutional neural networks perfor-mance[C]. International conference on hybrid artificial intelligence systems, Springer, Cham, 2017: 220-232.
[20] KRIZHEVSKY A, SUTSKEVER I, HINTON G E.ImageNet classification with deep convolutional neural networks[C]//International conference on neural information processing systems. Curran associates Inc. 2012: 1097-1105.
[21] Google brain. TensorFlow: A system for large-scale machine learning[C]. International conference on learning representations, 2016: 1-18.
[22] IOFFE S, SZEGEDY C.Batch normalization: Accelerating deep network training by reducing internal covariate shift[EB/OL].[2018-12-29].https://arxiv.org/abs/1502.03167.
[23] HE K, ZHANG X, REN S, et al.Deep residual learning for image recognition[C]. In proceedings of the IEEE conference on computer vision and pattern recognition, 2016: 770-778.
[24] REDMON J, FARHADI A.YOLO9000: Better, faster, stronger[C]. 2017 IEEE Conference on computer vision and pattern recognition (CVPR). USA: IEEE, 2017: 6517-6525.
[25] 杨观赐, 杨静, 李少波, 等. 基于Dopout与ADAM优化器的改进CNN算法[J]. 华中科技大学学报(自然科学版), 2018, 46(7):122-127.
YANG G, YANG J, LI S B, et al.Improved CNN algorithm based on dropout and ADAM optimizer[J]. Journal of Huazhong university of science and technology (Natural science edition), 2018, 46(7): 122-127.

基于ResNet-50深度卷积网络的果树病害智能诊断模型研究

The Intelligent Diagnosis Model of Fruit Tree Disease Based on ResNet-50

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 8

Metrics

本文评价

推荐阅读 0

[1]	余成斌. 文旅融合背景下乡村传统文化传承与保护策略研究[J]. 农业图书情报学报, 2021, 33(5): 51-63.
[2]	吴静, 曾桢. 农村地区信息贫困问题质性研究——以贵州省为例[J]. 农业图书情报学报, 2021, 33(4): 14-21.
[3]	张誉曜, 陈媛媛. 文本挖掘下的话语认知与构建——以人工智能和5G领域白宫新闻文本为例[J]. 农业图书情报学报, 2021, 33(4): 35-44.
[4]	安娜. 基于六何分析法的高校图书馆空间再造策略研究[J]. 农业图书情报学报, 2021, 33(4): 78-85.
[5]	梁晶璇, 宋慧琪, 陈佑成, 郑琼娥. 基于可视化分析的中国知识转移研究热点与展望[J]. 农业图书情报学报, 2021, 33(3): 38-55.
[6]	马祥涛. 文旅融合背景下图书馆文创产品开发探析[J]. 农业图书情报学报, 2020, 32(11): 15-21.
[7]	吕方婷. 文化IP视角下图书馆文化创意产品的开发研究[J]. 农业图书情报学报, 2020, 32(11): 22-29.
[8]	乔娜, 陈凯. 基于馆藏古籍文献资源的公共图书馆文创产品开发策略研究[J]. 农业图书情报学报, 2020, 32(11): 62-70.