果园机械化疏花技术与装备研究进展

doi:10.13733/j.jcam.issn.2095-5553.2024.10.050

摘要/Abstract

摘要： 为促进果园机械化疏花技术与装备发展，提高果园全程机械化水平，通过论述疏花的必要性与作业标准，以期为疏花工作提供理论指导。根据国内外手持式、振动式、机载式和智能式疏花机的工作方法与特点，概括分析其关键技术和优缺点。重点阐述机器视觉作为智能疏花机核心技术在花朵识别应用上的研究现状，通过对比YOLO、Faster R-CNN等典型花朵识别算法的平均精度、召回率和F1分数总结其制约因素和存在的主要问题。针对目前主流机载疏花机存在的工作模式单一、精准作业水平低、对果园标准化水平要求高、适用范围窄等主要问题，从规范果园种植方式、研发新型主轴结构与疏花绳材料、构建果园生产管理经验专家库、花朵识别技术的重点研究方向、智能疏花机未来研发重点5个方面进行展望。

关键词: 果园机械化, 疏花, 手持, 机载, 花朵识别

Abstract: In order to promote the development of mechanized thinning technology and equipment in orchards, improve the overall level of orchard mechanization, this article first discusses the necessity and operational standards of thinning flowers, with the aim of providing theoretical guidance for thinning work. According to domestic and foreign vibration‑type, handheld, airborne, and intelligent thinning machines working methods and characteristics, this article summarizes the advantages and disadvantages and analyzes the key technologies of the above equipment. It focuses on discussing the current research status of machine vision as the core support technology for intelligent thinning machines in flower recognition applications. By comparing typical flower recognition algorithms such as YOLO and Faster R-CNN, this article summarizes their limiting factors and major issues in terms of average precision, recall rate, and F1 score. In response to the main problems faced by current mainstream flower‑thinning machine, such as a single working mode, low precision level, high requirement for orchard standardization level, and narrow applicability range, this article proposes the following five prospects, such as the standardization of orchard planting methods, the development of new spindle structures and thinning rope materials, the construction of orchard production management experience expert database, key research directions of flower recognition technology, future research direction of intelligent flower thinning machine.

Key words: orchard mechanization, thinning flowers, in hand, airborne, flower recognition

中图分类号:

张振, 雷哓晖, 王伟, Andreas Herbst, 吕晓兰. 果园机械化疏花技术与装备研究进展[J]. 中国农机化学报, 2024, 45(10): 344-352.

Zhang Zhen, , Lei Xiaohui, Wang Wei, Andreas Herbst , Lü Xiaolan. Research progress of flower thinning technology and equipment in orchard mechanization[J]. Journal of Chinese Agricultural Mechanization, 2024, 45(10): 344-352.

参考文献

[ 1 ] 赵映, 肖宏儒, 梅松, 等. 我国果园机械化生产现状与发展策略[J]. 中国农业大学学报, 2017, 22(6): 116-127.
[ 2 ] 郑永军, 陈炳太, 吕昊暾, 等. 中国果园植保机械化技术与装备研究进展[J]. 农业工程学报, 2020, 36(20): 110-124.
Zheng Yongjun, Chen Bingtai, Lü Haotun, et al. Research progress of orchard plant protection mechanization technology and equipment in China [J]. Transactions of the Chinese Society of Agricultural Engineering, 2020, 36(20): 110-124.
[ 3 ] 傅锡敏, 吕晓兰, 丁为民, 等. 我国果园植保机械现状与技术需求[J]. 中国农机化学报, 2009(6): 10-13, 17.
Fu Ximin, Lü Xiaolan, Ding Weimin, et al. Present state and technical requirement about orchard plant protection machinery in China [J]. Journal of Chinese Agricultural Mechanization, 2009(6): 10-13, 17.
[ 4 ] 何雄奎. 中国精准施药技术和装备研究现状及发展建议[J]. 智慧农业(中英文), 2020, 2(1): 133-146.
[ 5 ] 潘云飞, 周艳, 何磊, 等. 果园管理工作中疏花疏果的研究进展[J]. 中国农机化学报, 2021, 42(11): 198-204.
Pan Yunfei, Zhou Yan, He Lei, et al. Research progress of flower and fruit thinning in orchard management [J]. Journal of Chinese Agricultural Mechanization, 2021, 42(11): 198-204.
[ 6 ] Kweon H J, Sagong D H. Influence of time of hand‑thining on bitter pit incidence and fruit quality in Gamhong/M. 9 apple trees [J]. Journal of Bio-Environment Control, 2021, 30(4): 342-350.
[ 7 ] Reighard G L, Souza F B M, Pio R. Peach cell number and size is affected by crop load and cultivar [C]. International Symposium on Modelling in Fruit Research and Orchard Management, 2015: 357-360.
[ 8 ] Purwantono A S D, Suparto S R. The influence of fruit thinning on fruit drop and quality of citrus [C]. IOP Conference Series: Earth and Environmental Science. IOP Publishing, 2019, 250(1): 012096.
[ 9 ] 薛晓敏, 王金政, 丛培建, 等. 苹果花果管理技术研究进展[J]. 江西农业学报, 2013, 25(12): 36-39, 43.
[10] Iwanami H, Moriya‑Tanaka Y, Honda C, et al. A model for representing the relationships among crop load, timing of thinning, flower bud formation, and fruit weight in apples [J]. Scientia Horticulturae, 2018, 242: 181-187.
[11] Iwanami H, Moriya‑Tanaka Y, Honda C, et al. Apple thinning strategy based on a model predicting flower‑bud formation [J]. Scientia Horticulturae, 2019, 256: 108529.
[12] 周吉生, 许春泽. 对苹果疏花定果技术的几点认识[J]. 果树实用技术与信息, 2015(6): 8.
[13] 唐德合. 苹果疏花疏果技术[J]. 果树资源学报, 2021, 2(4): 59-60.
[14] 赵书华. 苹果树疏花疏果“3步法”技术及应用效果[J]. 中国果树, 2010(5): 75.
[15] 俞晓红. 浅谈果树管理工作中常见问题及对策分析[J]. 农业技术与装备, 2020(3): 90-91.
[16] 聂佩显, 王金政, 路超, 等. 不同时期疏花疏果对红富士苹果花序坐果率和果实品质的影响[J]. 山东农业科学, 2013, 45(12): 27-29.
[17] Martín B, Torregrosa A, Brunton J G. Post‑bloom thinning of peaches for canning with hand‑held mechanical devices [J]. Scientia Horticulturae, 2010, 125(4): 658-665.
[18] 胡彩旗, 孙传海, 纪晶. 果树机械疏花机执行机构性能试验研究与分析[J]. 中国农机化学报, 2015, 36(5): 24-28.
Hu Caiqi, Sun Chuanhai, Ji Jing. Performance experiment and analysis on mechanical thinning flower actuator [J]. Journal of Chinese Agricultural Mechanization, 2015, 36(5): 24-28.
[19] 雷哓晖, 吕晓兰, 张美娜, 等. 电动甩绳式疏花机在翠冠梨园中的试验研究[J]. 中国农机化学报, 2020, 41(9): 47-52.
Lei Xiaohui, Lü Xiaolan, Zhang Meina, et al. Test and research on electric rotating rope blossom thinner in Cuiguan pear orchard [J]. Journal of Chinese Agricultural Mechanization, 2020, 41(9): 47-52.
[20] Martin‑Gorriz B, Torregrosa A, Brunton J G. Feasibility of peach bloom thinning with hand‑held mechanical devices [J]. Scientia Horticulturae, 2011, 129(1): 91-97.
[21] Martin‑Gorriz B, Torregrosa A, Brunton J G. Post‑bloom mechanical thinning for can peaches using a hand‑held electrical device [J]. Scientia Horticulturae, 2012, 144: 179-186.
[22] Schupp J R, Baugher T A, Miller S S, et al. Mechanical thinning of peach and apple trees reduces labor input and increases fruit size [J]. HortTechnology, 2008, 18(4): 660-670.
[23] Blanke M M, Damerow L. A novel device for precise and selective thinning in fruit crops to improve fruit quality [C]. International Symposium on Application of Precision Agriculture for Fruits and Vegetables, 2008: 275-280.
[24] Baugher T A, Ellis K, Remcheck J, et al. Mechanical string thinner reduces crop load at variable stages of bloom development of peach and nectarine trees [J]. HortScience, 2010, 45(9): 1327-1331.
[25] Steyn W J, Theron K I, De Villiers M H J. The effect of mechanical bloom thinning with the Darwin 300TM, on the hand thinning time, yield, and fruit quality in Zephyr nectarine [C]. XXIX International Horticultural Congress on Horticulture: Sustaining Lives, Livelihoods and Landscapes (IHC2014): 1130. 2014: 631-638.
[26] Solomakhin A A, Blanke M M. Mechanical flower thinning improves the fruit quality of apples [J]. Journal of the Science of Food and Agriculture, 2010, 90(5): 735-741.
[27] Assirelli A, Giovannini D, Cacchi M, et al. Evaluation of a new machine for flower and fruit thinning in stone fruits [J]. Sustainability, 2018, 10(11): 4088.
[28] Asteggiano L, Giordani L, Bevilacqua A, et al. Bloom mechanical thinning improves fruit quality and reduces production costs in peach [C]. VIII International Peach Symposium 1084. 2013: 389-394.
[29] Baugher T A, Schupp J, Ellis K, et al. String blossom thinner designed for variable tree forms increases crop load management efficiency in trials in four United States peach‑growing regions [J]. HortTechnology, 2010, 20(2): 409-414.
[30] 李君, 徐岩, 许绩彤, 等. 悬挂式电动柔性疏花机控制系统设计与试验[J]. 农业工程学报, 2016, 32(18): 61-66.
Li Jun, Xu Yan, Xu Jitong, et al. Design and experiment of control system for suspended electric flexible thinner [J]. Transactions of the Chinese Society of Agricultural Engineering, 2016, 32(18): 61-66.
[31] 李君, 陆华忠, 杨洲, 等. 荔枝柔性疏花器设计与试验研究[J]. 华南农业大学学报, 2013, 34(3): 431-435.
[32] 汪强. 矮密果树疏花装置设计及研究[D]. 保定: 河北农业大学, 2018.
[33] 雷哓晖, 吕晓兰, 张美娜, 等. 三节臂机载式疏花机的研制与试验[J]. 农业工程学报, 2019, 35(24): 31-38.
Lei Xiaohui, Lü Xiaolan, Zhang Meina, et al. Development and test of three arms tractor‑mounted flower thinner [J]. Transactions of the Chinese Society of Agricultural Engineering, 2019, 35(24): 31-38.
[34] Lyons D J, Heinemann P H, Schupp J R, et al. Development of a selective automated blossom thinning system for peaches [J]. Transactions of the ASABE, 2015, 58(6): 1447-1457.
[35] Reighard G L, Henderson W G. Mechanical blossom thinning in South Carolina peach orchards [C]. International Symposium on Mechanical Harvesting and Handling Systems of Fruits and Nuts 965. 2012: 117-121.
[36] Costa G, Blanke M M, Widmer A. Principles of thinning in fruit tree crops‑needs and novelties [C]. EUFRIN Thinning Working Group Symposia 998. 2012: 17-26.
[37] 吴升, 温维亮, 王传宇, 等. 数字果树及其技术体系研究进展[J]. 农业工程学报, 2021, 37(9): 350-360.
Wu Sheng, Wen Weiliang, Wang Chuanyu, et al. Research progress of digital fruit trees and its technology system [J]. Transactions of the Chinese Society of Agricultural Engineering, 2021, 37(9): 350-360.
[38] Wouters N, De Ketelaere B, Deckers T, et al. Multispectral detection of floral buds for automated thinning of pear [J]. Computers and Electronics in Agriculture, 2015, 113: 93-103.
[39] Horton R, Cano E, Bulanon D, et al. Peach flower monitoring using aerial multispectral imaging [J]. Journal of Imaging, 2017, 3(1): 2.
[40] Hung C, Nieto J, Taylor Z, et al. Orchard fruit segmentation using multi‑spectral feature learning [C]. 2013 IEEE/RSJ International Conference on Intelligent Robots and Systems. IEEE, 2013: 5314-5320.
[41] Bhattarai U, Bhusal S, Majeed Y, et al. Automatic blossom detection in apple trees using deep learning [J]. IFAC-PapersOnLine, 2020, 53(2): 15810-15815.
[42] Lin J, Li J, Yang Z, et al. Estimating litchi flower number using a multicolumn convolutional neural network based on a density map [J]. Precision Agriculture, 2022, 23(4): 1226-1247.
[43] Dias P A, Tabb A, Medeiros H. Apple flower detection using deep convolutional networks [J]. Computers in Industry, 2018, 99: 17-28.
[44] Dias P A, Tabb A, Medeiros H. Multispecies fruit flower detection using a refined semantic segmentation network [J]. IEEE Robotics and Automation Letters, 2018, 3(4): 3003-3010.
[45] Palacios F, Bueno G, Salido J, et al. Automated grapevine flower detection and quantification method based on computer vision and deep learning from on‑the‑go imaging using a mobile sensing platform under field conditions [J]. Computers and Electronics in Agriculture, 2020, 178: 105796.
[46] Wang X A, Tang J, Whitty M. Side‑view apple flower mapping using edge‑based fully convolutional networks for variable rate chemical thinning [J]. Computers and Electronics in Agriculture, 2020, 178: 105673.
[47] Ren S, He K, Girshick R, et al. Faster R-CNN: Towards real‑time object detection with region proposal networks [J]. Advances in Neural Information Processing Systems, 2015, 28.
[48] 沈跃, 李尚龙, 刘慧, 等. 基于Dog-Leg正则化自适应压缩采样的植株图像重构[J]. 农业工程学报, 2019, 35(12): 191-199.
Shen Yue, Li Shanglong, Liu Hui, et al. Plant image reconstruction based on Dog-Leg regularized adaptive compression sampling [J]. Transactions of the Chinese Society of Agricultural Engineering, 2019, 35(12): 191-199.
[49] 郭涛, 郭家, 李宗南, 等. 基于Darknet深度学习框架的桃花检测方法[J]. 中国农业信息, 2021, 33(6): 25-33.
[50] Bochkovskiy A, Wang C Y, Liao H Y M. Yolov4: Optimal speed and accuracy of object detection [J]. Arxiv Preprint Arxiv: 2004. 10934, 2020.
[51] 尚钰莹, 张倩如, 宋怀波. 基于YOLOv5s的深度学习在自然场景苹果花朵检测中的应用[J]. 农业工程学报, 2022, 38(9): 222-229.
Shang Yuying, Zhang Qqianru, Song Huaibo. Application of deep learning using YOLOv5s to apple flower detection in natural scenes [J]. Transactions of the Chinese Society of Agricultural Engineering, 2022, 38(9): 222-229.
[52] 邓颖, 吴华瑞, 朱华吉. 基于实例分割的柑橘花朵识别及花量统计[J]. 农业工程学报, 2020, 36(7): 200-207.
Deng Ying, Wu Huarui, Zhu Huaji. Recognition and counting of citrus flowers based on instance segmentation [J]. Transactions of the Chinese Society of Agricultural Engineering, 2020, 36(7): 200-207.
[53] 顾宝兴, 刘钦, 田光兆, 等. 基于改进YOLOv3的果树树干识别和定位[J]. 农业工程学报, 2022, 38(6): 122-129.
Gu Baoxing, Liu Qin, Tian Guangzhao, et al. Recognizing and locating the trunk of a fruit tree using improved YOLOv3 [J]. Transactions of the Chinese Society of Agricultural Engineering, 2022, 38(6): 122-129.
[54] 赵春江, 文朝武, 林森, 等. 基于级联卷积神经网络的番茄花期识别检测方法[J]. 农业工程学报, 2020, 36(24): 143-152.
Zhao Chunjiang, Wen Chaowu, Lin Sen, et al. Tomato florescence recognition and detection method based on cascaded neural network [J]. Transactions of the Chinese Society of Agricultural Engineering, 2020, 36(24): 143-152.
[55] 夏烨, 雷哓晖, 祁雁楠, 等. 基于改进Ghost-YOLOv5s-BiFPN算法检测梨树花序[J]. 智慧农业(中英文), 2022, 4(3): 108-119.

[1]	杨涛, 孙付春, 黄波, 吴柏强, 冉光泽. 果园作业平台关键技术研究进展[J]. 中国农机化学报, 2024, 45(1): 152-159.
[2]	周艳, 潘云飞, 何磊, 宋龙, 宋振帅, 王强. 牵引式疏花机设计与试验[J]. 中国农机化学报, 2022, 43(11): 43-49.
[3]	潘云飞, 周艳, 何磊, 王强, 宋振帅, 宋龙. 果园管理工作中疏花疏果的研究进展^*[J]. 中国农机化学报, 2021, 42(11): 198-204.
[4]	雷哓晖;吕晓兰;张美娜;马拯胞;常有宏;Andreas Herbst;. 电动甩绳式疏花机在翠冠梨园中的试验研究[J]. 中国农机化学报, 2020, 41(9): 47-52.
[5]	张紫涵;张红梅;宋月鹏;任龙龙;范国强;许令峰;. 偏置阶梯刀盘式果园开沟机设计与试验[J]. 中国农机化学报, 2020, 41(9): 53-56+75.
[6]	宋月鹏;张紫涵;范国强;高东升;张红梅;张晓辉;. 我国果园开沟施肥机械研究现状及发展趋势[J]. 中国农机化学报, 2019, 40(3): 7-12+25.