基于多尺度和注意力机制的番茄病害识别方法

doi:10.3969/j.issn.1004-1524.2021.07.19

浙江农业学报 ›› 2021, Vol. 33 ›› Issue (7): 1329-1338.DOI: 10.3969/j.issn.1004-1524.2021.07.19

基于多尺度和注意力机制的番茄病害识别方法

张宁¹^,²(), 吴华瑞²^,³^,⁴^,^*(), 韩笑²^,³^,⁴, 缪祎晟²^,³^,⁴

1.北京农学院计算机与信息工程学院,北京 102206
2.国家农业信息化工程技术研究中心,北京 100097
3.北京农业信息技术研究中心,北京 100097
4.农业农村部农业信息技术重点实验室,北京 100097

收稿日期:2020-09-25 出版日期:2021-07-25 发布日期:2021-08-06
通讯作者: 吴华瑞
作者简介:*吴华瑞,E-mail: wuhr@nercita.org.cn
张宁(1996—),女,山西永济人,硕士研究生,研究方向为农业信息化。E-mail: zhangning@nercita.org.cn
基金资助:
国家自然科学基金(61871041);国家大宗蔬菜产业技术体系岗位专家项目(CARS-23-C06);石家庄市科学技术研究与发展项目(201490074A)

Tomato disease recognition scheme based on multi-scale and attention mechanism

ZHANG Ning¹^,²(), WU Huarui²^,³^,⁴^,^*(), HAN Xiao²^,³^,⁴, MIAO Yisheng²^,³^,⁴

1. College of Computer and Information Engineering, Beijing University of Agriculture, Beijing 102206, China
2. National Engineering Research Center for Information Technology in Agriculture, Beijing 100097, China
3. Beijing Research Center for Information Technology in Agriculture, Beijing 100097, China
4. Key Laboratory of Agri-Informatics, Ministry of Agriculture and Rural Affairs, Beijing 100097, China

Received:2020-09-25 Online:2021-07-25 Published:2021-08-06
Contact: WU Huarui

摘要/Abstract

摘要：

番茄病害的及时发现与治理有助于提高番茄产量与质量,增加农户经济收益。利用物联网和人工智能可以无损害有效检测番茄病害,该研究提出了一种改进的AT-InceptionV3(Attention-InceptionV3)神经网络番茄叶部病害检测模型,该网络以InceptionV3为主干网络,结合多尺度卷积和注意力机制CBAM(convolutional block attention module,CBAM)模块,增强了病害信息表达并抑制无关信息干扰;同时引入迁移学习,防止样本数据量较少时出现过拟合的情况。为了评价优化模型的有效性,在Plant Village公开番茄病害数据集上进行了实验仿真测试。改进的模型在测试阶段对番茄健康叶片、细菌性斑疹病、晚疫病、叶霉病和黄曲病5种番茄常见叶片图像分类准确率达到98.4%,优化效果显著。为了进一步验证该方法在不同物联网中的普适性,实验对比了模型对不同分辨率病害图像的分类效果,结果表明,图像精度部分损失不会降低病害分类准确率。该模型能够为番茄温室智能网络决策判断提供重要依据。

关键词: 番茄, 多尺度卷积, 注意力机制, 迁移学习, 病害识别

Abstract:

The timely detection and management of tomato diseases can help increase the yield and quality of tomatoes and increase the economic benefits of farmers. The internet of things and artificial intelligence can effectively detect tomato diseases without damage. This paper proposed an improved AT-InceptionV3 (Attention-InceptionV3) neural network tomato leaf disease detection model. The network used InceptionV3 as the backbone network, combined with multi-scale convolution and attention mechanism CBAM (convolutional block attention module, CBAM) module, which enhanced disease information expression and suppressed irrelevant information interference. At the same time, transfer learning was introduced to prevent over-fitting when the sample data volume was small. In order to evaluate the effectiveness of the optimization model, an experimental simulation test was conducted on the public tomato disease data set of Plant Village. The improved model had an accuracy of 98.4% in classification of five leaf images of healthy tomato leaves, bacterial spot disease, late blight, leaf mold and yellow leaf curl vircus in the test stage, and the optimization effect was significant. In order to further verify the universality of this method in different Internet of Things, experiments had compared the model’s classification effect on disease images with different resolutions. The results showed that partial loss of image resolution would not reduce the accuracy of disease classification. This model could be used for tomato greenhouse intelligence and provide an important basis for network decision-making and judgment.

Key words: tomato, multi-scale convolution, attention mechanism, transfer learning, disease recognition

中图分类号:

张宁, 吴华瑞, 韩笑, 缪祎晟. 基于多尺度和注意力机制的番茄病害识别方法[J]. 浙江农业学报, 2021, 33(7): 1329-1338.

ZHANG Ning, WU Huarui, HAN Xiao, MIAO Yisheng. Tomato disease recognition scheme based on multi-scale and attention mechanism[J]. Acta Agriculturae Zhejiangensis, 2021, 33(7): 1329-1338.

图/表 11

参考文献 21

[1]	刘雪霞. 智慧农业与无线传感器应用分析[J]. 电子技术与软件工程, 2020(11):114-115.
	LIU X X. Smart agriculture and wireless sensor application analysis[J]. Electronic Technology & Software Engineering, 2020(11):114-115.(in Chinese)
[2]	张连宽, 肖德琴, 李就好. 基于无线传感器的作物图像传输方法[J]. 广东农业科学, 2013, 40(15):176-179.
	ZHANG L K, XIAO D Q, LI J H. Crop image transmission in wireless image sensor networks[J]. Guangdong Agricultural Sciences, 2013, 40(15):176-179.(in Chinese with English abstract)
[3]	李凯雨. 基于深度学习的农作物病害识别[D]. 郑州: 河南农业大学, 2018.
	LI K Y. The identification of crop diseases based on deep learning[D]. Zhengzhou: Henan Agricultural University, 2018. (in Chinese with English abstract)
[4]	SRIVASTAVA A, MA S, INOUE K. Development of a sensor for automatic detection of downey mildew disease [C]//International Conference on Intelligent Mechatronics and Automation, 2004:562-567.
[5]	王守志, 何东健, 李文, 等. 基于核K-均值聚类算法的植物叶部病害识别[J]. 农业机械学报, 2009, 40(3):152-155.
	WANG S Z, HE D J, LI W, et al. Plant leaf disease recognition based on kernel K-means clustering algorithm[J]. Transactions of the Chinese Society for Agricultural Machinery, 2009, 40(3):152-155.(in Chinese with English abstract)
[6]	殷建军, 潘春华, 肖克辉, 等. 基于无线图像传感器网络的农田远程监测系统[J]. 农业机械学报, 2017, 48(7):286-293.
	YIN J J, PAN C H, XIAO K H, et al. Remote monitoring system for farmland based on wireless image sensor network[J]. Transactions of the Chinese Society for Agricultural Machinery, 2017, 48(7):286-293.(in Chinese with English abstract)
[7]	申文韬. 5G通信技术与人工智能技术融合发展的基本现状与演化趋势[J]. 计算机产品与流通, 2020(7):38.
	SHEN W T. The basic status and evolutionary trend of the integration and development of 5G communication technology and artificial intelligence technology[J]. Computer Products and Circulation, 2020(7):38. (in Chinese)
[8]	郑远攀, 李广阳, 李晔. 深度学习在图像识别中的应用研究综述[J]. 计算机工程与应用, 2019, 55(12):20-36.
	ZHENG Y P, LI G Y, LI Y. Survey of application of deep learning in image recognition[J]. Computer Engineering and Applications, 2019, 55(12):20-36.(in Chinese with English abstract)
[9]	张帅. 基于深度学习的植物叶片识别算法研究[D]. 北京: 北京林业大学, 2016.
	ZHANG S. Research on plant leaf images identification algorithm based on deep learning[D]. Beijing: Beijing Forestry University, 2016. (in Chinese with English abstract)
[10]	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.(in Chinese with English abstract)
[11]	郭小清, 范涛杰, 舒欣. 基于改进Multi-Scale AlexNet的番茄叶部病害图像识别[J]. 农业工程学报, 2019, 35(13):162-169.
	GUO X Q, FAN T J, SHU X. Tomato leaf diseases recognition based on improved Multi-Scale AlexNet[J]. Transactions of the Chinese Society of Agricultural Engineering, 2019, 35(13):162-169.(in Chinese with English abstract)
[12]	吴华瑞. 基于深度残差网络的番茄叶片病害识别方法[J]. 智慧农业, 2019, 1(4):42-49.
	WU H R. Method of tomato leaf diseases recognition method based on deep residual network[J]. Smart Agriculture, 2019, 1(4):42-49.(in Chinese with English abstract)
[13]	SHAN C H, ZHANG J B, WANG Y J, et al. Attention-based end-to-end speech recognition on voice search [C]//2018 IEEE international conference on acoustics, speech and signal processing, 2018.
[14]	ADAM R K, ALEX B, INGMAR P. Hierarchical attentive recurrent tracking [C]//2017 IEEE conference on computer vision and pattern recognition, 2017.
[15]	CHU X, YANG W, OUYANG W, et al. Multi-context attention for human pose estimation [C]//2017 IEEE conference on computer vision and pattern recognition, 2017.
[16]	张盼盼, 李其申, 杨词慧. 基于轻量级分组注意力模块的图像分类算法[J]. 计算机应用, 2020, 40(3):645-650.
	ZHANG P P, LI Q S, YANG C H. Image classification algorithm based on lightweight group-wise attention module[J]. Journal of Computer Applications, 2020, 40(3):645-650.(in Chinese with English abstract)
[17]	袁建野, 南新元, 李成荣, 等. 注意力机制下的轻量级垃圾分类网络[J/OL]. 计算机工程, 2020. https://doi.org/10.19678/j.issn.1000-3428.0058108 .
	YUAN J Y, NAN X Y, LI C R, et al. Lightweight garbage classification network based on attention mechanism. Computer Engineering, 2020. https://doi.org/10.19678/j.issn.1000-3428.0058108 . (in Chinese with English abstract)
[18]	胡志伟, 杨华, 黄济民, 等. 基于注意力残差机制的细粒度番茄病害识别[J]. 华南农业大学学报, 2019, 40(6):124-132.
	HU Z W, YANG H, HUANG J M, et al. Fine-grained tomato disease recognition based on attention residual mechanism[J]. Journal of South China Agricultural University, 2019, 40(6):124-132.(in Chinese with English abstract)
[19]	WOO S, PARK J, LEE J Y, et al. CBAM: convolutional block attention module[M]//Computer Vision-ECCV 2018. Cham: Springer International Publishing, 2018: 3-19.
[20]	DENG J, DONG W, SOCHER R, et al. ImageNet: a large-scale hierarchical image database[C]//2009 IEEE Conference on Computer Vision and Pattern Recognition. June 20-25, 2009, Miami, FL, USA. IEEE, 2009: 248-255.
[21]	林朝剑, 张广群, 杨洁, 等. 基于迁移学习的林业业务图像识别[J]. 南京林业大学学报(自然科学版), 2020, 44(4):215-221.
	LIN C J, ZHANG G Q, YANG J, et al. Transfer learning based recognition for forestry business images[J]. Journal of Nanjing Forestry University (Natural Sciences Edition), 2020, 44(4):215-221.(in Chinese with English abstract)

网络层数 Number of network layers	类型 Type	滤波器 Filter shape	输入尺寸 Input size
1	卷积层1 Convolutional layer 1	3×3	299×299×3
2	卷积层2 Convolutional layer 2	3×3	149×149×32
3	卷积层3 Convolutional layer 3	3×3	147×147×32
4	池化层1 Pooling layer 1	3×3	147×147×64
5	卷积层4 Convolutional layer 4	3×3	73×73×64
6	卷积层5 Convolutional layer 5	3×3	73×73×80
7	卷积层6 Convolutional layer 6	3×3	35×35×192
8	Inception1模块组 Inception1 module group	3个Inception1 3 Inception1	35×35×288
9	Inception2模块组 Inception2 module group	5个Inception2 5 Inception2	17×17×768
10	Inception3模块组 Inception3 module group	3个Inception3 3 Inception3	8×8×1 280
11	CBAM模块 CBAM module	7×7	8×8×2 048
12	全连接层 Fully connected layer	分类器 Classifier	1×2 048

网络层数 Number of network layers	类型 Type	滤波器 Filter shape	输入尺寸 Input size
1	卷积层1 Convolutional layer 1	3×3	299×299×3
2	卷积层2 Convolutional layer 2	3×3	149×149×32
3	卷积层3 Convolutional layer 3	3×3	147×147×32
4	池化层1 Pooling layer 1	3×3	147×147×64
5	卷积层4 Convolutional layer 4	3×3	73×73×64
6	卷积层5 Convolutional layer 5	3×3	73×73×80
7	卷积层6 Convolutional layer 6	3×3	35×35×192
8	Inception1模块组 Inception1 module group	3个Inception1 3 Inception1	35×35×288
9	Inception2模块组 Inception2 module group	5个Inception2 5 Inception2	17×17×768
10	Inception3模块组 Inception3 module group	3个Inception3 3 Inception3	8×8×1 280
11	CBAM模块 CBAM module	7×7	8×8×2 048
12	全连接层 Fully connected layer	分类器 Classifier	1×2 048

病害类型 Types of disease	InceptionV3从零训练 InceptionV3 training from zero		InceptionV3迁移学习 InceptionV3 transfer learning		AT-InceptionV3迁移学习 AT-InceptionV3 transfer learning
病害类型 Types of disease	准确率 Accuracy	召回率 Recall	准确率 Accuracy	召回率 Recall	准确率 Accuracy	召回率 Recall
细菌性斑疹病Bacterial spot	58.88	56.60	96.30	93.90	93.75	97.40
健康叶Healthy leaf	87.50	74.50	100.0	97.60	100.0	96.38
晚疫病Late blight	61.30	64.50	95.00	98.70	98.75	98.75
叶霉病Leaf mold	62.50	67.60	96.30	96.30	98.75	98.75
黄曲病Yellow leaf curl virus	65.00	71.20	97.50	98.70	98.75	98.75

病害类型 Types of disease	InceptionV3从零训练 InceptionV3 training from zero		InceptionV3迁移学习 InceptionV3 transfer learning		AT-InceptionV3迁移学习 AT-InceptionV3 transfer learning
病害类型 Types of disease	准确率 Accuracy	召回率 Recall	准确率 Accuracy	召回率 Recall	准确率 Accuracy	召回率 Recall
细菌性斑疹病Bacterial spot	58.88	56.60	96.30	93.90	93.75	97.40
健康叶Healthy leaf	87.50	74.50	100.0	97.60	100.0	96.38
晚疫病Late blight	61.30	64.50	95.00	98.70	98.75	98.75
叶霉病Leaf mold	62.50	67.60	96.30	96.30	98.75	98.75
黄曲病Yellow leaf curl virus	65.00	71.20	97.50	98.70	98.75	98.75

原标签 True lable		晚期病害Late disease				早期病害Early diease
原标签 True lable		图片 Images		预测标签与置信度 Predict label and confidence degree		图片 Images	预测标签与置信度 Predict label and confidence degree
番茄细菌性斑疹病 Tomato bacterial spot				番茄细菌性斑疹病(score=0.972 39) Tomato bacterial spot (score=0.972 39) 番茄叶霉病(score=0.014 88) Tomato leaf mold (score=0.014 88) 番茄黄曲病(score=0.007 67) Tomato yellow leaf curl virus (score=0.007 67) 番茄晚疫病(score=0.003 56) Tomato late blight (score=0.003 56) 番茄健康叶(score=0.001 49) Tomato healthy leaf(score=0.001 49)			番茄细菌性斑疹病(score=0.941 05) Tomato bacterial spot (score=0.941 05) 番茄健康叶(score=0.027 93) Tomato healthy leaf(score=0.027 93) 番茄叶霉病(score=0.014 04) Tomato leaf mold (score=0.014 04) 番茄黄曲病(score=0.012 65) Tomato yellow leaf curl virus (score=0.012 65) 番茄晚疫病(score=0.004 32) Tomato late blight (score=0.004 32)
番茄健康叶 Tomato healthy leaf				番茄健康叶(score=0.832 44) Tomato healthy leaf(score=0.832 44) 番茄晚疫病(score=0.081 10) Tomato late blight (score=0.081 10) 番茄叶霉病(score=0.080 07) Tomato leaf mold (score=0.080 07) 番茄黄曲病(score=0.004 26) Tomato yellow leaf curl virus (score=0.004 26) 番茄细菌性斑疹病(score=0.002 13) Tomato bacterial spot (score=0.002 13)			番茄健康叶(score=0.918 93) Tomato healthy leaf(score=0.918 93) 番茄晚疫病(score=0.070 37) Tomato late blight (score=0.070 37) 番茄细菌性斑疹病(score=0.006 04) Tomato bacterial spot (score=0.006 04) 番茄黄曲病(score=0.003 07) Tomato yellow leaf curl virus (score=0.003 07) 番茄叶霉病(score=0.001 60) Tomato leaf mold (score=0.001 60)
番茄晚疫病 Tomato late blight				番茄晚疫病(score=0.999 18) Tomato late blight (score=0.999 18) 番茄细菌性斑疹病(score=0.000 56) Tomato bacterial spot (score=0.000 56) 番茄叶霉病(score=0.000 11) Tomato leaf mold (score=0.000 11) 番茄健康叶(score=0.000 10) Tomato healthy leaf(score=0.000 10) 番茄黄曲病(score=0.000 05) Tomato yellow leaf curl virus (score=0.000 05)			番茄晚疫病(score=0.621 03) Tomato late blight (score=0.621 03) 番茄细菌性斑疹病(score=0.139 23) Tomato bacterial spot (score=0.139 23) 番茄健康叶(score=0.130 01) Tomato healthy leaf(score=0.130 01) 番茄黄曲病(score=0.089 71) Tomato yellow leaf curl virus (score=0.089 71) 番茄叶霉病(score=0.020 02) Tomato leaf mold (score=0.020 02)
番茄叶霉病 Tomato leaf mold				番茄叶霉病(score=0.994 35) Tomato leaf mold (score=0.994 35) 番茄晚疫病(score=0.002 57) Tomato late blight (score=0.002 57) 番茄黄曲病(score=0.001 33) Tomato yellow leaf curl virus (score=0.001 33) 番茄健康叶(score=0.001 05) Tomato healthy leaf (score=0.001 05) 番茄细菌性斑疹病(score=0.000 70) Tomato bacterial spot (score=0.000 70)			番茄叶霉病(score=0.992 87) Tomato leaf mold (score=0.992 87) 番茄细菌性斑疹病(score=0.004 54) Tomato bacterial spot (score=0.004 54) 番茄晚疫病(score=0.001 39) Tomato late blight (score=0.001 39) 番茄黄曲病(score=0.001 09) Tomato yellow leaf curl virus (score=0.001 09) 番茄健康叶(score=0.000 11) Tomato healthy leaf(score=0.000 11)
原标签 True lable	晚期病害Late disease				早期病害Early diease
原标签 True lable	图片 Images		预测标签与置信度 Predict label and confidence degree		图片 Images		预测标签与置信度 Predict label and confidence degree
番茄黄曲病 Tomato yellow leaf curl virus			番茄黄曲病(score=0.993 05) Tomato yellow leaf curl virus (score=0.993 05) 番茄细菌性斑疹病(score=0.003 78) Tomato bacterial spot (score=0.003 78) 番茄晚疫病(score=0.001 20) Tomato late blight (score=0.001 20) 番茄叶霉病(score=0.001 20) Tomato leaf mold (score=0.001 20) 番茄健康叶(score=0.000 77) Tomato healthy leaf(score=0.000 77)				番茄黄曲病(score=0.975 95) Tomato yellow leaf curl virus (score=0.975 95) 番茄叶霉病(score=0.008 98) Tomato leaf mold (score=0.008 98) 番茄健康叶(score=0.008 70) Tomato healthy leaf(score=0.008 70) 番茄细菌性斑疹病(score=0.005 33) Tomato bacterial spot (score=0.005 33) 番茄晚疫病(score=0.001 04) Tomato late blight (score=0.001 04)

基于多尺度和注意力机制的番茄病害识别方法

Tomato disease recognition scheme based on multi-scale and attention mechanism

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 11

参考文献 21

相关文章 15

编辑推荐

Metrics

本文评价

模型 Model	迁移学习 Transfer learning	CBAM模块 CBAM module	准确率 Accuracy/%	运行时间 Training time/s
InceptionV3从零训练 InceptionV3 training from zero			67.8	93 000
InceptionV3迁移学习 InceptionV3 transfer learning	√		97.7	1 966
AT-InceptionV3迁移学习 AT-InceptionV3 transfer learning	√	√	98.4	2 031

图像分辨率 Images resolution	单个病害识别准确率Single disease recognition accuracy					平均准确率 Average accuracy
图像分辨率 Images resolution	细菌性斑疹病 Bacterial spot	健康叶 Healthy leaf	晚疫病 Late blight	叶霉病 Leaf mold	黄曲病 Yellow leaf curl virus	平均准确率 Average accuracy
256×256	93.75	100.00	98.75	98.75	98.75	98.00
240×240	99.00	97.00	99.00	97.00	100.00	98.40
224×224	99.00	97.00	98.00	98.00	100.00	98.40
128×128	99.00	100.00	98.00	99.00	100.00	99.20

原标签 True lable	分辨率 Resolution	预测标签与置信度 Predict label and confidence degree	预测标签与置信度 Predict label and confidence degree
番茄细菌性斑疹病 Tomato bacterial spot	256×256	番茄黄曲病(score=0.492 34) Tomato yellow leaf curl virus (score=0.492 34) 番茄细菌性斑疹病(score=0.438 02) Tomato bacterial spot (score=0.438 02) 番茄健康叶(score=0.030 68) Tomato healthy leaf(score=0.030 68) 番茄叶霉病(score=0.030 27) Tomato leaf mold (score=0.030 27) 番茄晚疫病(score=0.008 68) Tomato late blight (score=0.008 68)	番茄晚疫病(score=0.513 22) Tomato late blight (score=0.513 22) 番茄细菌性斑疹病(score=0.393 29) Tomato bacterial spot (score=0.393 29) 番茄叶霉病(score=0.086 98) Tomato leaf mold (score=0.086 98) 番茄黄曲病(score=0.005 80) Tomato yellow leaf curl virus (score=0.005 80) 番茄健康叶(score=0.000 70) Tomato healthy leaf(score=0.000 70)
番茄细菌性斑疹病 Tomato bacterial spot	240×240	番茄黄曲病(score=0.649 63) Tomato yellow leaf curl virus(score=0.649 63) 番茄细菌性斑疹病(score=0.288 60) Tomato bacterial spot (score=0.288 60) 番茄健康叶(score=0.032 27) Tomato healthy leaf(score=0.032 27) 番茄叶霉病(score=0.017 99) Tomato leaf mold (score=0.017 99) 番茄晚疫病(score=0.011 51) Tomato late blight (score=0.011 51)	番茄细菌性斑疹病(score=0.684 45) Tomato bacterial spot (score=0.684 45) 番茄晚疫病(score=0.242 76) Tomato late bligh (score=0.242 76) 番茄叶霉病(score=0.067 28) Tomato leaf mold (score=0.067 28) 番茄黄曲病(score=0.004 97) Tomato yellow leaf curl virus (score=0.004 97) 番茄健康叶(score=0.000 53) Tomato healthy leaf (score=0.000 53)
番茄细菌性斑疹病 Tomato bacterial spot	224×224	番茄黄曲病(score=0.640 14) Tomato yellow leaf curl virus (score=0.640 14) 番茄细菌性斑疹病(score=0.309 77) Tomato bacterial spot (score=0.309 77) 番茄健康叶(score=0.023 06) Tomato healthy leaf(score=0.023 06) 番茄叶霉病(score=0.017 49) Tomato leaf mold (score=0.017 49) 番茄晚疫病(score=0.009 55) Tomato late blight (score=0.009 55)	番茄细菌性斑疹病(score=0.768 61) Tomato bacterial spot (score=0.768 61) 番茄晚疫病(score=0.160 56) Tomato late blight (score=0.160 56) 番茄叶霉病(score=0.064 90) Tomato leaf mold (score=0.064 90) 番茄黄曲病(score=0.005 26) Tomato yellow leaf curl virus (score=0.005 26) 番茄健康叶(score=0.000 66) Tomato healthy leaf (score=0.000 66)
番茄细菌性斑疹病 Tomato Bacterial spot	128×128	番茄细菌性斑疹病(score=0.491 46) Tomato bacterial spot (score=0.491 46) 番茄黄曲病(score=0.475 77) Tomato yellow leaf curl virus (score=0.475 77) 番茄叶霉病(score=0.017 26) Tomato leaf mold (score=0.017 26) 番茄晚疫病(score=0.010 84) Tomato late blight (score=0.010 84) 番茄健康叶(score=0.004 68) Tomato healthy leaf (score=0.004 68)	番茄细菌性斑疹病(score=0.623 09) Tomato bacterial spot (score=0.623 09) 番茄晚疫病(score=0.270 19) Tomato late bligh (score=0.270 19) 番茄叶霉病(score=0.093 92) Tomato leaf mold (score=0.093 92) 番茄黄曲病(score=0.012 38) Tomato yellow leaf curl virus (score=0.012 38) 番茄健康叶(score=0.000 43) Tomato healthy leaf (score=0.000 43)

[1]	赵虎, 张越亭, 刘永华. 糖分含量对番茄叶片Pst DC3000抗性的影响及其机理[J]. 浙江农业学报, 2021, 33(6): 1001-1011.
[2]	王克磊, 朱隆静, 苏世闻, 包玉花, 陈先知, 徐坚. 不同规格穴盘对番茄幼苗生长及其机械化移栽的影响[J]. 浙江农业学报, 2021, 33(5): 840-845.
[3]	王见, 田光宝, 周勤. 基于迁移学习的棉花识别[J]. 浙江农业学报, 2020, 32(8): 1457-1465.
[4]	王同林, 叶红霞, 郑积荣, 李明. 番茄果实中主要风味物质研究进展[J]. 浙江农业学报, 2020, 32(8): 1513-1522.
[5]	乔亚丽, 郁继华, 李旺雄, 金宁, 金莉, 吕剑, 肖雪梅, 唐中褀, 胡琳莉. 番茄果实不同采收时间矿质元素变化分析[J]. 浙江农业学报, 2020, 32(3): 430-436.
[6]	刘新宇, 陈鹏, 张光辉, 赵俊杰, 李博浩, 张南, 孙韦珂, 许娟娟, 叶行涛, 魏金鹏, 于高波. 外源脯氨酸对番茄体内残留百菌清降解的调控作用[J]. 浙江农业学报, 2020, 32(3): 437-446.
[7]	张芳, 王佩欣, 何勇, 骆慧枫, 寿国忠. 基于物联网的阳台微型温室作物生长环境因子探究[J]. 浙江农业学报, 2020, 32(2): 234-242.
[8]	李乐, 田敏娇, 高艳明, 李建设. 硒肥对基质培番茄生长和矿质元素积累的影响[J]. 浙江农业学报, 2020, 32(2): 253-261.
[9]	朱晓林, 魏小红, 王宝强, 王贤, 张明君. c-GMP诱导对盐胁迫下番茄的转录组分析[J]. 浙江农业学报, 2020, 32(10): 1788-1797.
[10]	孙德智, 杨恒山, 张庆国, 范富, 苏雅乐其其格, 彭靖, 韩晓日. 外源一氧化氮供体硝普钠对番茄幼苗盐胁迫伤害的缓解作用[J]. 浙江农业学报, 2019, 31(8): 1286-1294.
[11]	杨亚娜, 樊小雪, 徐刚, 张宇, 李亚灵, 温祥珍. 不同红蓝LED光照强度和灌溉量交互作用对番茄幼苗生长的影响[J]. 浙江农业学报, 2019, 31(5): 737-745.
[12]	林薇, 周海霞, 兰挚谦, 张凯歌, 刘吉青, 张雪艳. 基于葵花杆硫酸铵生物基肥的番茄不同生育期配方肥的效果[J]. 浙江农业学报, 2019, 31(5): 756-765.
[13]	邢鲲, 曹俊宇, 赵飞. 设施番茄昆虫群落组成及时间动态的聚类分析[J]. 浙江农业学报, 2019, 31(4): 600-606.
[14]	王晓敏, 赵宇飞, 袁东升, 刘珮君, 郑福顺, 胡新华, 付金军, 高艳明, 李建设. 三十三个番茄自交系数量性状的配合力和遗传力分析[J]. 浙江农业学报, 2019, 31(12): 2025-2035.
[15]	朱隆静, 陈先知, 周友和, 苏世闻, 王克磊, 徐坚. 潮汐灌溉在番茄育苗上的应用研究[J]. 浙江农业学报, 2018, 30(2): 242-247.