基于深度学习与颜色特征规则的茶树炭疽病病斑量化分析

黄洁琼; 邓卓然; 任恒泽; 吕务云; 陆梦倩; 王新超; 陈雅楠; 王玉春

doi:10.13305/j.cnki.jts.2026.03.007

茶叶科学 >

2026 , Vol. 46 >Issue 3: 475 - 488

DOI: https://doi.org/10.13305/j.cnki.jts.2026.03.007

研究报告

基于深度学习与颜色特征规则的茶树炭疽病病斑量化分析

黄洁琼 ,
邓卓然 ,
任恒泽 ,
吕务云 ,
陆梦倩 ,
王新超 ,
陈雅楠 ,
王玉春

展开

1.浙江农林大学茶学与茶文化学院/浙江省植物种质资源保育与利用国际科技合作基地,浙江杭州 311300;
2.中国农业科学院茶叶研究所/国家茶树改良中心/农业农村部特种经济动植物生物学与遗传育种重点实验室,浙江杭州 310008

黄洁琼,女,硕士研究生,从事茶树抗病机理研究。

^*yananchen@zafu.edu.cn;ycwang0201@126.com

收稿日期: 2025-12-12

修回日期: 2026-01-27

网络出版日期: 2026-06-16

基金资助

浙江省“三农九方”科技协作计划（2025SNJF036）; 国家重点研发计划（2024YFD1200504）

收起

Quantitative Analysis of Tea Anthracnose Lesions Based on Deep Learning and Color-Feature Rules

HUANG Jieqiong ,
DENG Zhuoran ,
REN Hengze ,
LÜ Wuyun ,
LU Mengqian ,
WANG Xinchao ,
CHEN Yanan ,
WANG Yuchun

Expand

1. College of Tea Science and Tea Culture, Zhejiang A & F University / International Science and Technology Cooperation Base for Conservation and Utilization of Plant Germplasm Resources of Zhejiang Province, Hangzhou 311300, China;
2. Tea Research Institute, Chinese Academy of Agricultural Sciences / National Center for Tea Improvement / Key Laboratory of Biology, Genetics and Breeding of Special Economic Animals and Plants, Ministry of Agriculture and Rural Affairs, P. R. China, Hangzhou 310008, China

Received date: 2025-12-12

Revised date: 2026-01-27

Online published: 2026-06-16

Fold

摘要

炭疽病是危害茶树生长发育和茶叶品质的主要叶部病害之一。目前国内尚未建立茶树对该病害的抗性鉴定标准,常规鉴定评价工作主要借鉴其他作物,且相关评价方法依赖人工目测,造成评价方法主观性强、评价工作效率较低和评价结果准确性较弱等问题,严重制约了茶树良种选育工作进度和规范性。基于深度学习和图像识别技术,首先利用掩膜区域卷积神经网络（Mask R-CNN）实现田间病叶的实例分割与定位;随后在叶片感兴趣区域（ROI）内融合HSV颜色空间阈值规则与绿色度指数（Excess green,ExG）进行病斑精细分割,并根据病斑面积占比构建0~9级分级标准,开发了自动化评价系统。结果显示,模型在叶片实例检测上取得叶片掩膜分割交并比（mIoU）为92.56%,IoU阈值为0.5时的平均精度（R_AP@0.5）为98.91%;在病斑区域检测一致性验证中,自动化测量的病斑面积比例与Fiji-Weka工具生成的人工标注值极显著相关（Pearson’s r=0.946,P<0.001）,达到了可替代人工主观评价病害分级标准的精度要求。

关键词： 茶树炭疽病; 抗性评价; 掩膜区域卷积神经网络; 颜色特征规则

本文引用格式

黄洁琼 , 邓卓然 , 任恒泽 , 吕务云 , 陆梦倩 , 王新超 , 陈雅楠 , 王玉春 . 基于深度学习与颜色特征规则的茶树炭疽病病斑量化分析[J]. 茶叶科学, 2026 , 46(3) : 475 -488 . DOI: 10.13305/j.cnki.jts.2026.03.007

Abstract

Anthracnose is one of the major foliar diseases that threatens the growth and development of tea plants and compromises tea quality. At present, no standardized criterion has been established in China for evaluating tea resistance to this disease. Conventional identification and evaluation practices are largely adapted from other crops, and the associated methods rely on visual inspection, resulting in strong subjectivity, low efficiency and limited accuracy. These shortcomings severely constrain both the progress and standardization of elite tea cultivar breeding. In this study, we developed an automated evaluation framework based on deep learning and image analysis. First, Mask R-CNN was used to localize and perform instance segmentation of diseased leaves from field images. Subsequently, within the leaf region of interest (ROI), fine lesion segmentation was achieved by integrating HSV color-space thresholding rules with the Excess Green (ExG) index. Based on the lesion area ratio, a 0-9 severity grading standard was established, and an automated evaluation system was developed. The results show that the model achieved mean of intersection over union (mIoU) of 92.56% for leaf mask segmentation and ratio of average precision at IoU=0.5 (R_AP@0.5) of 98.91% for leaf instance detection. For consistency validation of lesion detection, the lesion area ratios measured by the automated method were highly significantly correlated with manual reference values generated using the Fiji-Weka tool (Pearson’s r=0.946, P<0.001), meeting the accuracy requirement to replace subjective manual grading.

Key words： tea anthracnose; resistance evaluation; Mask R-CNN; color-feature rules

参考文献

[1] Chen L, Apostolides Z, Chen Z M.Global tea breeding[M]. Hangzhou: Zhejiang University Press & Springer, 2012: 1-2.
[2] Wang Y C, Hao X Y, Wang L, et al. Diverse Colletotrichum species cause anthracnose of tea plants (Camellia sinensis (L.) O. Kuntze) in China[J]. Scientific Reports, 2016, 6: 35287. https://doi.org/10.1038/srep35287.
[3] 刘威, 袁丁, 郭桂义, 等. 茶树炭疽病病原鉴定[J]. 南方农业学报, 2017, 48(3): 448-453.
Liu W, Yuan D, Guo G Y, et al.Identification of anthracnose pathogen in tea plant[J]. Journal of Southern Agriculture, 2017, 48(3): 448-453.
[4] 陈亮, 杨亚军, 虞富莲. 中国茶树种质资源研究的主要进展和展望[J]. 植物遗传资源学报, 2004, 5(4): 389-392.
Chen L, Yang Y J, Yu F L.Tea germplasm research in China: recent progresses and prospects[J]. Journal of Plant Genetic Resources, 2004, 5(4): 389-392.
[5] 吴鹏飞, 龚洪恩, 姚小华, 等. 普通油茶无性系抗炭疽病评价[J]. 林业科学研究, 2018, 31(4): 158-163.
Wu P F, Gong H E, Yao X H, et al.Evaluating the resistance of Camellia oleifera clones to anthracnose[J]. Forest Research, 2018, 31(4): 158-163.
[6] Wang L, Wang Y C, Cao H L, et al. Transcriptome analysis of an anthracnose-resistant tea plant cultivar reveals genes associated with resistance to Colletotrichum camelliae[J]. PLoS ONE, 2016, 11(2): e0148535. https://doi.org/10.1371/journal.pone.0148535.
[7] Bock C H, Chiang K S, Del Ponte E M. Plant disease severity estimated visually: a century of research, best practices,opportunities for improving methods and practices to maximize accuracy[J]. Tropical Plant Pathology, 2022, 47: 25-42. https://doi.org/10.1080/07352681003617285.
[8] 武伟, 汪璇, 王成秋, 等. 作物病害图像识别研究进展——基于文献计量的分析[J]. 西南大学学报(自然科学版), 2024, 46(10): 42-54.
Wu W, Wang X, Wang C Q, et al.Research progress on crop disease image recognition: a bibliometric analysis[J]. Journal of Southwest University (Natural Science Edition), 2024, 46(10): 42-54.
[9] Lu Y, Yi S J, Zeng N, et al.Identification of rice diseases using deep convolutional neural networks[J]. Neurocomputing, 2017, 267: 378-384. https://doi.org/10.1016/j.neucom.2017.06.023.
[10] Picon A, Seitz M, Alvarez-Gila A, et al.Crop conditional Convolutional Neural Networks for massive multi-crop plant disease classification over cell phone acquired images taken on real field conditions[J]. Computers and Electronics in Agriculture, 2019, 167: 105093. https://doi.org/10.1016/j.compag.2019.105093.
[11] Rajeena P P F, Aswathy S U, Moustafa M A, et al. Detecting plant disease in corn leaf using EfficientNet architecture: an analytical approach[J]. Electronics, 2023, 12(8): 1938. https://doi.org/10.3390/electronics12081938.
[12] Fuentes A, Yoon S, Kim S C, et al.A robust deep-learning-based detector for real-time tomato plant diseases and pests recognition[J]. Sensors, 2017, 17(9): 2022. doi: 10.3390/s17092022.
[13] He K M, Gkioxari G, Dollár P, et al.Mask R-CNN[C]//2017 IEEE International Conference on Computer Vision (ICCV), Venice: IEEE, 2017: 2980-2988. https://doi.org/10.1109/ICCV.2017.322.
[14] Ren S Q, He K M, Girshick R, et al.Faster R-CNN: towards real-time object detection with region proposal networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(6): 1137-1149.
[15] Lin T Y, Dollár P, Girshick R, et al.Feature pyramid networks for object detection[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu: IEEE, 2017: 936-944. https://doi.org/10.1109/CVPR.2017.106.
[16] Li X D, Zhou Y H, Liu J Y, et al.The detection method of potato foliage diseases in complex background based on instance segmentation and semantic segmentation[J]. Frontiers in Plant Science, 2022, 13: 899754. https://doi.org/10.3389/fpls.2022.899754.
[17] Liu B Y, Fan K J, Su W H, et al. Two-stage convolutional neural networks for diagnosing the severity of alternaria leaf blotch disease of the apple tree[J]. Remote Sensing, 2022, 14(11): 2519. https://doi.org/10.3390/rs14112519.
[18] Hamuda E, Glavin M, Jones E.Automatic crop detection under field conditions using the HSV colour space and morphological operations[J]. Computers and Electronics in Agriculture, 2017, 133: 97-107. https://doi.org/10.1016/j.compag.2016.11.021.
[19] Guo M, Pan Y M, Dai Y L, et al. First report of brown blight disease caused by Colletotrichum gloeosporioides on Camellia sinensis in Anhui Province, China[J]. Plant Disease, 2014, 98(2): 284. https://doi.org/10.1094/PDIS-08-13-0896-PDN.
[20] Jeyaraj A, Elango T, Chen X, et al.Advances in understanding the mechanism of resistance to anthracnose and induced defence response in tea plants[J]. Molecular Plant Pathology. 2023, 24(10): 1330-1346.
[21] Otsu N.A threshold selection method from gray-level histograms[J]. IEEE Transactions on Systems, Man, and Cybernetics, 1979, 9(1): 62-66.
[22] 冈萨雷斯, 伍兹. 数字图像处理[M]. 3版. 阮秋琦, 译. 北京: 电子工业出版社, 2011: 407-408.
Gonzalez R C, Woods R E.Digital image processing[M]. 3rd ed. Ruan Q Q translated. Beijing: Publishing House of Electronics Industry, 2011: 407-408.
[23] Ghiasi G, Cui Y, Srinivas A, et al.Simple copy-paste is a strong data augmentation method for instance segmentation[C]//2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Nashville: IEEE, 2021: 2917-2927. https://doi.org/10.1109/CVPR46437.2021.00294.
[24] Pavicic M, Overmyer K, Rehman A U, et al.Image-based methods to score fungal pathogen symptom progression and severity in excised Arabidopsis leaves[J]. Plants, 2021, 10(1): 158. doi: 10.3390/plants10010158.
[25] 李红莉, 周铁锋, 毛宇骁, 等. 西湖龙井产区茶树炭疽菌分离鉴定及其植物源抑菌剂筛选[J]. 茶叶科学, 2023, 43(2): 194-204.
Li H L, Zhou T F, Mao Y X, et al.Isolation and identification of anthracnose pathogen from Xihu Longjing plantation and screening of its plant-derived fungicides[J]. Journal of Tea Science, 2023, 43(2): 194-204.
[26] Chen W L, Ye T, Sun Q Y, et al.Arbuscular mycorrhizal fungus alleviates anthracnose disease in tea seedlings[J]. Frontiers in Plant Science, 2023, 13: 1058092. https://doi.org/10.3389/fpls.2022.1058092.
[27] Yoshida K, Takeda Y.Evaluation of anthracnose resistance among tea genetic resources by wound-inoculation assay[J]. Japan Agricultural Research Quarterly (JARQ), 2006, 40(4): 379-386.
[28] 张孟婷, 李学慧, 陈虹雨, 等. 不同茶树品种对茶树炭疽菌的抗性快筛初探[J]. 茶业通报, 2022, 44(2): 76-80.
Zhang M T, Li X H, Chen H Y, et al.Preliminary study on rapid screening of resistance of different tea plant varieties to Colletotrichum camelliae[J]. Journal of Tea Business, 2022, 44(2): 76-80.
[29] Ferentinos K P.Deep learning models for plant disease detection and diagnosis[J]. Computers and Electronics in Agriculture, 2018, 145: 311-318. https://doi.org/10.1016/j.compag.2018.01.009.
[30] Khan A I, Quadri S M K, Banday S, et al. Deep diagnosis: a real-time apple leaf disease detection system based on deep learning[J]. Computers and Electronics in Agriculture, 2022, 198: 107093. https://doi.org/10.1016/j.compag.2022.107093.
[31] Wang C S, Du P F, Wu H R, et al.A cucumber leaf disease severity classification method based on the fusion of DeepLabV3+ and U-Net[J]. Computers and Electronics in Agriculture, 2021, 189: 106373. https://doi.org/10.1016/j.compag.2021.106373.
[32] Tassis L M, De Souza J E T, Krohling R A. A deep learning approach combining instance and semantic segmentation to identify diseases and pests of coffee leaves from in-field images[J]. Computers and Electronics in Agriculture, 2021, 186: 106191. https://doi.org/10.1016/j.compag.2021.106191.
[33] Xue Z Y, Xu R J, Bai D, et al. YOLO-Tea: a tea disease detection model improved by YOLOv5[J]. Forests, 2023, 14(2): 415. https://doi.org/10.3390/f14020415.
[34] Hu G S, Wei K, Zhang Y, et al. Estimation of tea leaf blight severity in natural scene images[J]. Precision Agriculture, 2021, 22: 1239-1254. https://doi.org/10.1007/s11119-020-09782-8.
[35] Sural S, Qian G, Pramanik S.Segmentation and histogram generation using the HSV color space for image retrieval[C]//Proceedings of the International Conference on Image Processing. Rochester: IEEE, 2002, 2: 589-592. https://doi.org/10.1109/ICIP.2002.1040019.
[36] Müller-Linow M, Wilhelm J, Briese C, et al. Plant Screen Mobile: an open-source mobile device app for plant trait analysis[J]. Plant Methods, 2019, 15: 2. https://doi.org/10.1186/s13007-019-0386-z.
[37] Waldamichael F G, Debelee T G, Schwenker F, et al.Coffee disease detection using a robust HSV color-based segmentation and transfer learning for use on smartphones[J]. International Journal of Intelligent Systems, 2022, 37(8): 4967-4993.

Options

文章导航

模态框（Modal）标题

摘要

本文引用格式

Abstract

参考文献