基于机器视觉的小贯小绿叶蝉智能识别的研究与应用

doi:10.13305/j.cnki.jts.20220506.001

茶叶科学 ›› 2022, Vol. 42 ›› Issue (3): 376-386.doi: 10.13305/j.cnki.jts.20220506.001

基于机器视觉的小贯小绿叶蝉智能识别的研究与应用

边磊¹, 何旭栋^2,3, 季慧华², 蔡晓明¹, 罗宗秀¹, 陈华才^3,*, 陈宗懋^1,*

1.中国农业科学院茶叶研究所,浙江杭州 310008;
2.杭州益昊农业科技有限公司,浙江杭州 310018;
3.中国计量大学,浙江杭州 310018

收稿日期:2021-10-09 修回日期:2021-12-01 出版日期:2022-06-15 发布日期:2022-06-17
通讯作者: *huacaichen@cjlu.edu.cn;zmchen2006@163.com
作者简介:边磊,男,副研究员,主要从事茶树害虫物理防治技术研究。
基金资助:
浙江重点研发计划（2019C02033）、财政部和农业农村部：国家现代农业产业技术体系（CARS-19）、中国农业科学院创新工程

Research and Application of Intelligent Identification of Empoasca onukii Based on Machine Vision

BIAN Lei¹, HE Xudong^2,3, JI Huihua², CAI Xiaoming¹, LUO Zongxiu¹, CHEN Huacai^3,*, CHEN Zongmao^1,*

1. Tea Research Institute, Chinese Academy of Agricultural Sciences, Hangzhou 310008, China;
2. Hangzhou Yihao Agricultural Technology Co., Ltd., Hangzhou 310018, China;
3. China Jiliang University, Hangzhou 310018, China

Received:2021-10-09 Revised:2021-12-01 Online:2022-06-15 Published:2022-06-17

摘要/Abstract

摘要： 深度学习已经在农作物害虫实时监测的智能识别过程中广泛应用。以小贯小绿叶蝉（Empoasca onukii）识别模型为基础,研究深度学习在诱虫板上叶蝉识别中的应用,旨在提高小贯小绿叶蝉田间种群调查的准确性。本研究设计了一种茶园小贯小绿叶蝉的识别、计数方法,首先采用黄色诱虫板诱集小贯小绿叶蝉,利用相机对诱虫板进行图像采集并上传至服务器,然后通过服务器部署的目标检测算法,对图像中叶蝉进行识别与计数。通过算法筛选,确定YOLOv3作为识别算法,用改进后的Soft-NMS代替原来的NMS,用K-means聚类方法计算新的先验框的尺寸,提升YOLOv3对目标识别的速度和准确率。通过田间试验对比诱虫板上叶蝉的真实数量,结果显示优化后识别算法的准确率可达到95.35%以上。本研究验证了诱虫板诱集、目标识别算法和物联网技术相结合,能够为小贯小绿叶蝉田间种群的实时监测提供技术支持,可为其他具有颜色偏爱性昆虫的实时监测和茶园害虫综合治理提供参考。

关键词: 深度学习, 目标检测, 小贯小绿叶蝉, 种群监测, YOLOv3

Abstract: Deep learning has been widely used in intelligent identification in the progress of real-time monitoring of crop pests. Based on the identification model of tea leafhopper, Empoasca onukii, the application of deep learning in field leafhopper recognition was introduced to improve the precision of field population investigation of E. onukii. In this paper, a method of identification and count of E. onukii in tea garden was designed. Firstly, yellow sticky card was used to attract tea leafhoppers, and images of cards were collected by camera and uploaded to the web server. Then, target detection algorithm deployed by the server was used to identify and count the leafhoppers in the images. Through algorithm screening, YOLOv3 was determined as the recognition algorithm, and the improved soft-NMS was used to replace the original NMS. K-means clustering method was used to calculate the size of the new prior frame, so as to improve the speed and precision of YOLOv3. The results show that the average precision of the optimized algorithm could reach more than 95.35% comparing with the real number of leafhoppers on the sticky card. Therefore, the combination of the sticky card trapping, target recognition algorithm and internet of things technology could realize the real-time monitoring of population for E. onukii, which could provide a reference for other insects with color preference and integrated pest management in tea gardens.

Key words: deep learning, target detection, Empoasca onukii, population monitoring, YOLOv3

中图分类号:

边磊, 何旭栋, 季慧华, 蔡晓明, 罗宗秀, 陈华才, 陈宗懋. 基于机器视觉的小贯小绿叶蝉智能识别的研究与应用[J]. 茶叶科学, 2022, 42(3): 376-386. doi: 10.13305/j.cnki.jts.20220506.001.

BIAN Lei, HE Xudong, JI Huihua, CAI Xiaoming, LUO Zongxiu, CHEN Huacai, CHEN Zongmao. Research and Application of Intelligent Identification of Empoasca onukii Based on Machine Vision[J]. Journal of Tea Science, 2022, 42(3): 376-386. doi: 10.13305/j.cnki.jts.20220506.001.

参考文献

[1] Kogan M.Integrated pest management: historical perspectives and contemporary developments[J]. Annual Review of Entomology, 1998, 43(1): 243-270.
[2] 吴孔明, 陆宴辉, 王振营. 我国农业害虫综合防治研究现状与展望[J]. 昆虫知识, 2009, 46(6): 831-836.
Wu K M, Lu Y H, Wang Z Y.Advance in integrated pest management of crops in China[J]. Chinese Bulletin of Entomology, 2009, 46(6): 831-836.
[3] 中国茶叶学会标准化工作委员会. 茶树主要害虫绿色防控技术规程: T/CTSS 37—2021 [S/OL]. [2021-12-20]. http://down.foodmate.net/standard/yulan.php?itemid=115353.
Standardization Committee of China Tea Science Society. Green controlling technical regulations for major pests of tea plant: T/CTSS 37—2021 [S/OL]. [2021-12-20]. http://down.foodmate.net/standard/yulan.php?itemid=115353.
[4] Khanna A, Kaur S.Evolution of internet of things (IoT) and its significant impact in the field of precision agriculture[J]. Computers and Electronics in Agriculture, 2019, 157: 218-231.
[5] Bian L, Sun X L, Luo Z X, et al.Design and selection of trap color for capture of the tea leafhopper, Empoascavitis, by orthogonal optimization[J]. Entomologia Experimentalis et Applicata, 2014, 151(3): 247-258.
[6] Bian L, Cai X M, Luo Z X, et al.Design of an attractant for Empoasca onukii (Hemiptera: Cicadellidae) based on the volatile components of fresh tea leaves[J]. Journal of Economic Entomology, 2018, 111(2): 629-636.
[7] 封洪强, 姚青. 农业害虫自动识别与监测技术[J]. 植物保护, 2018, 44(5): 127-133.
Feng H Q, Yao Q.Automatic identification and monitoring technology of agricultural pests[J]. Plant Protection, 2018, 44(5): 127-133.
[8] 姚青, 吴叔珍, 蒯乃阳, 等. 基于改进CornerNet的水稻灯诱飞虱自动检测方法构建与验证[J]. 农业工程学报, 2021, 37(7): 183-189.
Yao Q, Wu S Z, Kuai N Y, et al.Automatic detection of rice planthoppers through light-trap insect images using improved Corner Net[J]. Transactions of the Chinese Society of Agricultural Engineering, 2021, 37(7): 183-189.
[9] 王茂林, 荣二花, 张利军, 等. 基于图像处理的蓟马计数方法研究[J]. 山西农业科学, 2020, 48(5): 812-816.
Wang M L, Rong E H, Zhang L J, et al, Study on counting Thripidae thysanoptera based on image processing[J]. Journal of Shanxi Agricultural Sciences, 2020, 48(5): 812-816.
[10] 李正飞. 基于小波变换的图像增强技术研究[J]. 机械工程与自动化, 2009(2): 15-17.
Li Z F.Image enhancement method based on wavelet transform[J]. Mechanical Engineering & Automation, 2009(2): 15-17.
[11] 张江勇. 基于深度学习的动车关键部位故障图像检测[D]. 成都: 电子科技大学, 2019.
Zhang J Y.Fault detection of emus based on deep learning [D]. Chengdu: University of Electronic Science and Technology of China, 2019.
[12] 彭宜. 基于残差网络和随机森林的音频识别方法研究[D]. 武汉: 武汉科技大学, 2019.
Peng Y.Research on audio recognition method based on residual network and random forest [D]. Wuhan: Wuhan University of Science and Technology, 2019.
[13] Girshick R.Fast R-CNN[C]//IEEE. 2015 IEEE International Conference on Computer Vision (ICCV). Santiago: 2015.
[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] Liu W, Anguelov D, Erhan D, et al.SSD: single shot multibox detector[M]//Leibe B, Matas J, Sebe N, et al. Lecture notes in computer science. Cham: Springer, 2016: 21-37.
[16] Redmon J, Farhadi A.YOLOv3: an incremental improvement[J]. Eprint ArXiv, 2018: 1804.02767. doi: 10.48550/arXiv.1804.02767.
[17] Huang J, Zhang H, Wang L, et al.Improved YOLOv3 Model for miniature camera detection[J]. Optics and Laser Technology, 2021, 142. doi: 10.1016/j.optlastec.2021.107133.
[18] Bodla N, Singh B, Chellappa R, et al.Soft-NMS-improving object detection with one line of code[J]. Eprint ArXiv, 2017: 1704.04503. doi: 10.48550/arXiv.1704.04503.
[19] Redmon J, Farhadi A.YOLO9000: better, faster, stronger[C]//IEEE. 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu: 2017.
[20] 边磊, 孙晓玲, 陈宗懋. 假眼小绿叶蝉的日飞行活动性及成虫飞行能力的研究[J]. 茶叶科学, 2014, 34(3): 248-252.
Bian L, Sun X L, Chen Z M.Studies on daily flight activity and adult flight capacity of Empoasca vitis Göthe[J]. Journal of Tea Science, 2014, 34(3): 248-252.
[21] Kim K N, Huang Q Y, Lei C L.Advances in insect phototaxis and application to pest management: a review[J]. Pest Management Science, 2019, 75(12): 3135-3143.
[22] Shimoda M, Honda K.Insect reactions to light and its applications to pest management[J]. Applied Entomology and Zoology, 2013, 48(4): 413-421.
[23] Rodriguez-Saona C R, Byers J A, Schiffhauer D. Effect of trap color and height on captures of blunt-nosed and sharp-nosed leafhoppers (Hemiptera: Cicadellidae) and non-target arthropods in cranberry bogs[J]. Crop Protection, 2012, 40: 132-144.
[24] 杨智辉. 黄色诱虫板对茶果园主要害益虫的诱杀作用调查[J]. 现代农业科技, 2017(15): 91-93.
Yang Z H.Investigation on the trapping and killing effect of yellow insect traps on the main harmful and beneficial insects in tea orchards[J]. Modern Agricultural Science and Technology, 2017(15): 91-93.
[25] 吴孔明. 中国农作物病虫害防控科技的发展方向[J]. 农学学报, 2018, 8(1): 35-38.
Wu K M.Development direction of crop pest control science and technology in China[J]. Journal of Agriculture, 2018, 8(1): 35-38.
[26] 陈宗懋, 蔡晓明, 周利, 等. 中国茶园有害生物防控40年[J]. 中国茶叶, 2020, 42(1): 1-8.
Chen Z M, Cai X M, Zhou L, et al.Developments on tea plant pest control in past 40 years in China[J]. China Tea, 2020, 42(1): 1-8.

基于机器视觉的小贯小绿叶蝉智能识别的研究与应用

Research and Application of Intelligent Identification of Empoasca onukii Based on Machine Vision

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 9

Metrics

本文评价

推荐阅读 10

[1]	郭明明, 李兆群, 刘岩, 饶福强, 俞嘉伟, 吴鲁超, 周利, 陈宗懋. 双丙环虫酯对小贯小绿叶蝉的防治效果及残留评价[J]. 茶叶科学, 2022, 42(3): 358-366.
[2]	张怡, 赵珠蒙, 王校常, 冯海强, 林杰. 基于ResNet卷积神经网络的绿茶种类识别模型构建[J]. 茶叶科学, 2021, 41(2): 261-271.
[3]	任倩倩, 庄明珠, 蔡晓明, 边磊, 罗宗秀, 李兆群, 尤民生, 陈宗懋, 金珊. 小贯小绿叶蝉取食诱导抗、感茶树品种挥发物的释放[J]. 茶叶科学, 2020, 40(6): 795-806.
[4]	刘丰静, 冉伟, 李喜旺, 汪素琴, 孙晓玲. 小贯小绿叶蝉在5个茶树品种（系）上的蜜露排泄量与茶树叶片结构比较[J]. 茶叶科学, 2020, 40(5): 625-631.
[5]	于永晨, 肖斌, 孙晓玲. 小贯小绿叶蝉半胱氨酸蛋白酶基因的克隆与表达[J]. 茶叶科学, 2018, 38(3): 263-270.
[6]	王奇志, 刘育梅, 李书明. 入侵植物胜红蓟精油对小贯小绿叶蝉的忌避和熏蒸活性研究[J]. 茶叶科学, 2017, 37(5): 442-448.
[7]	郑雨婷, 韩善捷, 韩宝瑜. 基于感官智能分析仪器的蝉害茶色香味的变异分析[J]. 茶叶科学, 2017, 37(5): 449-457.
[8]	王瑶, 慕卫, 张丽霞, 许永玉, 林琎, 张正群. 杀虫剂对茶园3种常见刺吸式口器害虫的室内毒力评价[J]. 茶叶科学, 2017, 37(4): 392-398.
[9]	王薛婷, 韩宝瑜. 交尾致雌雄叶蝉搜寻叶片和色彩行为差异的视频轨迹分析[J]. 茶叶科学, 2016, 36(5): 544-550.