基于三维点云和集成学习的大田烟草株型特征识别

doi:10.3785/j.issn.1008-9209.2021.05.173

浙江大学学报（农业与生命科学版）

2022, Vol. 48

Issue (3): 393-402 DOI: 10.3785/j.issn.1008-9209.2021.05.173

农业工程

基于三维点云和集成学习的大田烟草株型特征识别

贾奥博¹(

),董天浩¹,张彦²,朱冰琳¹,孙延国²,吴元华²,石屹²,马韫韬¹,郭焱¹(

)

^1.中国农业大学土地科学与技术学院，北京 100193
^2.中国农业科学院烟草研究所，山东青岛 266101

Recognition of field-grown tobacco plant type characteristics based on three-dimensional point cloud and ensemble learning

Aobo JIA¹(

),Tianhao DONG¹,Yan ZHANG²,Binglin ZHU¹,Yanguo SUN²,Yuanhua WU²,Yi SHI²,Yuntao MA¹,Yan GUO¹(

)

^1.College of Land Science and Technology, China Agricultural University, Beijing 100193, China
^2.Institute of Tobacco Research, Chinese Academy of Agricultural Sciences, Qingdao 266101, Shandong, China

全文: PDF(2228 KB) HTML

摘要：

为构建高效的大田烟草株型定量化方法，本研究基于多视角图像序列并采用运动恢复结构算法重建了5个品种烟草植株的三维点云。根据常用的烟草株型特征指标，基于烟株三维点云自动提取株高、顶宽、底宽、叶层最大宽等10个表型参数，并基于大田原位手动测量的株高和叶层最大宽对计算精度进行评估。结果表明，基于三维点云提取的株高和叶层最大宽与实测值的决定系数（R²）均大于0.97，均方根误差分别为3.0、3.1 cm。同时，采用不同方法对提取的烟草表型性状进行分析。组间相关性分析结果表明，有16对性状呈极显著正相关，1对性状呈极显著负相关。单因素多元方差分析结果表明，各品种株型之间具有极显著差异。利用主成分分析提取前3个主成分，其对总体方差的累计贡献率为81.6%。基于Stacking集成学习方法进行株型判别，其准确率达到93.7%，显著高于随机森林、支持向量机和朴素贝叶斯等3种机器学习模型的准确率。本研究可为大田烟草表型特征及株型识别提供方法依据。

关键词： 三维点云; 烟草; 表型; 机器学习; 株型

Abstract:

To develop an efficient method for quantifying tobacco plant types in the field, the three-dimensional (3D) point clouds of individual plant of five tobacco cultivars were reconstructed based on multi-view image sequences using the structure from motion method. According to the plant type characteristic indexes commonly used, ten phenotypic parameters such as plant height, top width, bottom width, and maximum width of leaf layer were automatically extracted based on the 3D point cloud of tobacco plant, and the calculation accuracy was evaluated based on the plant height and maximum width of leaf layer measured manually in situ in the field. The results indicated the coefficients of determination (R²) of the plant height and maximum width of leaf layer extracted from the 3D point cloud were all greater than 0.97, and the root mean square errors were 3.0, 3.1 cm, respectively. Meanwhile, the extracted phenotypic parameters of tobacco plants were analyzed by different methods. The results of intergroup correlation analysis showed that 16 pairs of traits were extremely significant positive correlations, while one pair of traits was extremely significant negative correlation. The results of one-way multivariate analysis of variance showed that there were highly significant differences among the plant types. The first three principal components were extracted by principal component analysis, and their cumulative contribution rate to the overall variance was 81.6%. The accuracy of plant type discrimination was 93.7% using Stacking ensemble learning method, which was significantly higher than those using random forest, support vector machine and naive Bayesian. This study can provide a method basis for phenotypic characteristics and plant type recognition of field-grown tobacco plants.

Key words: three-dimensional point cloud tobacco phenotype machine learning plant type

收稿日期: 2021-05-17 出版日期: 2022-07-07

CLC:

TP 391.7

基金资助: 中国烟草总公司山东省公司重点项目“山东‘中棵烟’特征及形成机制研究”(201803)

通讯作者: 郭焱 E-mail: aobo.jia@cau.edu.cn;yan.guo@cau.edu.cn

作者简介: 贾奥博（https://orcid.org/0000-0003-2657-9607），E-mail：aobo.jia@cau.edu.cn

	服务
	把本文推荐给朋友
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	贾奥博
	董天浩
	张彦
	朱冰琳
	孙延国
	吴元华
	石屹
	马韫韬
	郭焱

引用本文:

贾奥博,董天浩,张彦,朱冰琳,孙延国,吴元华,石屹,马韫韬,郭焱. 基于三维点云和集成学习的大田烟草株型特征识别[J]. 浙江大学学报（农业与生命科学版）, 2022, 48(3): 393-402.

Aobo JIA,Tianhao DONG,Yan ZHANG,Binglin ZHU,Yanguo SUN,Yuanhua WU,Yi SHI,Yuntao MA,Yan GUO. Recognition of field-grown tobacco plant type characteristics based on three-dimensional point cloud and ensemble learning. Journal of Zhejiang University (Agriculture and Life Sciences), 2022, 48(3): 393-402.

链接本文:

https://www.zjujournals.com/agr/CN/10.3785/j.issn.1008-9209.2021.05.173 或 https://www.zjujournals.com/agr/CN/Y2022/V48/I3/393

图1 基于图像序列的烟草点云重建及预处理

表1 提取的烟草表型参数

图2 烟草各表型参数提取示意图A.烟株正视图；B.最小包围盒体积及凸包体积；C.烟株俯视图；D.茎叶夹角。各表型参数的含义见表1，下同。

图 3 基于Stacking的集成学习方法

图4 基于多视角图像序列重建的5个烟草品种植株的冠层点云HH：红花大金元；XK：新K326；LY：辽烟1号；QG：青梗；EMS：EMS突变体。下同。

图5 基于冠层重建点云计算的烟草株高、叶层最大宽与测量值的比较n：样本数。

图6 表型参数相关性分析*和**分别表示在P＜0.05和P＜0.01水平显著和极显著相关。

图 7 烟草表型对各主成分的贡献率

图 8 烟草品种间各表型的差异比较短栅上不同小写字母表示在P＜0.05水平差异有统计学意义

表2 不同模型的株型判别准确率及评价指标比较

1	DONALD C M. The breeding of crop ideotypes[J]. Euphytica, 1968, 17(3): 385-403. DOI:10.1007/bf00056241 doi: 10.1007/bf00056241
2	袁隆平.超级杂交水稻育种研究的进展[J].中国稻米,2008(1):1-3. YUAN L P. Study development on breeding of super hybrid rice[J]. China Rice, 2008(1): 1-3. (in Chinese)
3	王丰,丁伟,冯勇刚,等.烤烟优质适产理想株型探讨[J].种子,2007,26(5):84-87. WANG F, DING W, FENG Y G, et al. Study on tobacco ideotype for high quality and adequate production[J]. Seed, 2007, 26(5): 84-87. (in Chinese)
4	顾会战,母明新,史洪涛,等.关于烤烟“中棵烟”培育的若干思考[J].中国烟草学报,2020,26(6):89-96. DOI:10.16472/j.chinatobacco.2020.T0015 GU H Z, MU M X, SHI H T, et al. Some thoughts on 'Zhongkeyan'‍ cultivation of flue-cured tobacco[J]. Acta Tabacaria Sinica, 2020, 26(6): 89-96. (in Chinese with English abstract) doi: 10.16472/j.chinatobacco.2020.T0015
5	赵春江.植物表型组学大数据及其研究进展[J].农业大数据学报,2019,1(2):5-18. DOI:10.19788/j.issn.2096-6369.190201 ZHAO C J. Big data of plant phenomics and its research progress[J]. Journal of Agricultural Big Data, 2019, 1(2): 5-18. (in Chinese with English abstract) doi: 10.19788/j.issn.2096-6369.190201
6	郭焱,史同鑫,吴劼,等.烟草植株静态虚拟模型的研究[J].中国烟草学报,2012,18(5):29-33. DOI:10.3969/j.issn.1004-5708.2012.05.005 GUO Y, SHI T X, WU J, et al. Development of static virtual tobacco model based on three-dimensional scanning methodology[J]. Acta Tabacaria Sinica, 2012, 18(5): 29-33. (in Chinese with English abstract) doi: 10.3969/j.issn.1004-5708.2012.05.005
7	王芸芸,温维亮,郭新宇,等.烟草地上部植株三维重构与可视化[J].中国农业科学,2013,46(1):37-44. DOI:10.3864/j.issn.0578-1752.2013.01.005 WANG Y Y, WEN W L, GUO X Y, et al. Research on three-dimensional reconstruction and visualization of above ground tobacco plant[J]. Scientia Agricultura Sinica, 2013, 46(1): 37-44. (in Chinese with English abstract) doi: 10.3864/j.issn.0578-1752.2013.01.005
8	DUAN T, CHAPMAN S C, HOLLAND E, et al. Dynamic quantification of canopy structure to characterize early plant vigour in wheat genotypes[J]. Journal of Experimental Botany, 2016, 67(15): 4523-4534. DOI:10.1093/jxb/erw227 doi: 10.1093/jxb/erw227
9	HUI F, ZHU J, HU P, et al. Image-based dynamic quantification and high-accuracy 3D evaluation of canopy structure of plant populations[J]. Annals of Botany, 2018, 121(5): 1079-1088. DOI:10.1093/aob/mcy016 doi: 10.1093/aob/mcy016
10	朱冰琳,刘扶桑,朱晋宇,等.基于机器视觉的大田植株生长动态三维定量化研究[J].农业机械学报,2018,49(5):256-262. DOI:10.6041/j.issn.1000-1298.2018.05.030 ZHU B L, LIU F S, ZHU J Y, et al. Three-dimensional quantifications of plant growth dynamics in field-grown plants based on machine vision method[J]. Transactions of the Chinese Society for Agricultural Machinery, 2018, 49(5): 256-262. (in Chinese with English abstract) doi: 10.6041/j.issn.1000-1298.2018.05.030
11	HOU T Y, ZHENG B Y, XU Z L, et al. Simplification of leaf surfaces from scanned data: effects of two algorithms on leaf morphology[J]. Computers and Electronics in Agriculture, 2016, 121: 393-403. DOI:10.1016/j.compag.2016.01.010 doi: 10.1016/j.compag.2016.01.010
12	MOHANTY S P, HUGHES D P, SALATHÉ M. Using deep learning for image-based plant disease detection[J]. Frontiers in Plant Science, 2016, 7: 1419. DOI:10.3389/fpls.2016.01419 doi: 10.3389/fpls.2016.01419
13	KUSSUL N, LAVRENIUK M, SKAKUN S, et al. Deep learning classification of land cover and crop types using remote sensing data[J]. IEEE Geoscience and Remote Sensing Letters, 2017, 14(5): 778-782. DOI:10.1109/LGRS.2017.2681128 doi: 10.1109/LGRS.2017.2681128
14	CHLINGARYAN A, SUKKARIEH S, WHELAN B. Machine learning approaches for crop yield prediction and nitrogen status estimation in precision agriculture: a review[J]. Computers and Electronics in Agriculture, 2018, 151: 61-69. DOI:10.1016/j.compag.2018.05.012 doi: 10.1016/j.compag.2018.05.012
15	CHEN T, AOIKE T, YAMASAKI M, et al. Predicting rice heading date using an integrated approach combining a machine learning method and a crop growth model[J]. Frontiers in Genetics, 2020, 11: 599510. DOI:10.3389/fgene. 2020.599510 doi: 10.3389/fgene. 2020.599510
16	LI B, ZHANG N, WANG Y G, et al. Genomic prediction of breeding values using a subset of SNPs identified by three machine learning methods[J]. Frontiers in Genetics, 2018, 9: 237. DOI:10.3389/fgene.2018.00237 doi: 10.3389/fgene.2018.00237
17	袁培森,杨承林,宋玉红,等.基于Stacking集成学习的水稻表型组学实体分类研究[J].农业机械学报,2019,50(11):144-152. DOI:10.6041/j.issn.1000-1298.2019.11.016 YUAN P S, YANG C L, SONG Y H, et al. Classification of rice phenomics entities based on Stacking ensemble learning[J]. Transactions of the Chinese Society for Agricultural Machinery, 2019, 50(11): 144-152. (in Chinese with English abstract) doi: 10.6041/j.issn.1000-1298.2019.11.016
18	WANG J, XU J, PENG Y, et al. Prediction of forest unit volume based on hybrid feature selection and ensemble learning[J]. Evolutionary Intelligence, 2020, 13(1): 21-32. DOI:10.1007/s12065-019-00219-4 doi: 10.1007/s12065-019-00219-4
19	鲁冬冬,邹进贵.三维激光点云的降噪算法对比研究[J].测绘通报,2019():102-105. DOI:10.13474/j.cnki.11-2246.2019.0599 LU D D, ZOU J G. Comparative research on denoising algorithms of 3D laser point cloud[J]. Bulletin of Surveying and Mapping, 2019(): 102-105. (in Chinese with English abstract) doi: 10.13474/j.cnki.11-2246.2019.0599
20	FISCHLER M A, BOLLES R C. Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography[J]. Communications of the ACM, 1981, 24(6): 381-395. DOI:10.1145/358669.358692 doi: 10.1145/358669.358692
21	XIAO S F, CHAI H H, SHAO K, et al. Image-mased dynamic quantification of aboveground structure of sugar beet in field[J]. Remote Sensing, 2020, 12(2): 269. DOI:10.3390/rs12020269 doi: 10.3390/rs12020269
22	薛小平,赵会纳,陈懿,等.贵州烟区烤烟K326株型特征研究[J].中国烟草科学,2013,34(1):34-39. DOI:10.3969/j.issn.1007-5119.2013.01.007 XUE X P, ZHAO H N, CHEN Y, et al. Studies on plant type characteristics of flue-cured tobacco K326 in Guizhou[J]. Chinese Tobacco Science, 2013, 34(1): 34-39. (in Chinese with English abstract) doi: 10.3969/j.issn.1007-5119.2013.01.007
23	国家烟草专卖局. 烟草农艺性状调查测量方法: [S].北京:中国标准出版社,2010. DOI:10.1136/tc.2010.036079 State Tobacco Monopoly Administration. Investigating and Measuring Methods of Agronomical Character of Tobacco: [S]. Beijing: Standards Press of China, 2010. (in Chinese) doi: 10.1136/tc.2010.036079
24	孔彦龙,高晓阳,李红玲,等.基于机器视觉的马铃薯质量和形状分选方法[J].农业工程学报,2012,28(17):143-148. DOI:10.3969/j.issn.1002-6819.2012.17.021 KONG Y L, GAO X Y, LI H L, et al. Potato grading method of mass and shapes based on machine vision[J]. Transactions of the CSAE, 2012, 28(17): 143-148. (in Chinese with English abstract) doi: 10.3969/j.issn.1002-6819.2012.17.021
25	吴正敏,曹成茂,王二锐,等.基于形态特征参数的茶叶精选方法[J].农业工程学报,2019,35(11):315-321. DOI:10.11975/j.issn.1002-6819.2019.11.036 WU Z M, CAO C M, WANG E R, et al. Tea selection method based on morphology feature parameters[J]. Transactions of the CSAE, 2019, 35(11): 315-321. (in Chinese with English abstract) doi: 10.11975/j.issn.1002-6819.2019.11.036
26	柴宏红,邵科,于超,等.基于三维点云的甜菜根表型参数提取与根型判别[J].农业工程学报,2020,36(10):181-188. DOI:10.11975/j.issn.1002-6819.2020.10.022 CHAI H H, SHAO K, YU C, et al. Extraction of phenotypic parameters and discrimination of beet root types based on 3D point cloud[J]. Transactions of the CSAE, 2020, 36(10): 181-188. (in Chinese with English abstract) doi: 10.11975/j.issn.1002-6819.2020.10.022
27	孙统,漆建波,黄华国.手持式激光雷达观测玉兰物候期叶倾角变化[J].遥感信息,2020,35(5):113-118. DOI:10.3969/j.issn.1000-3177.2020.05.014 SUN T, QI J B, HUANG H G. Using handheld LiDAR to observe leaf inclination angels of Magnolia denudatain phenological period[J]. Remote Sensing Information, 2020, 35(5): 113-118. (in Chinese with English abstract) doi: 10.3969/j.issn.1000-3177.2020.05.014
28	WOLD S, ESBENSEN K, GELADI P. Principal component analysis[J]. Chemometrics and Intelligent Laboratory Systems, 1987, 2(1/2/3): 37-52. DOI:10.1016/0169-7439(87)80084-9 doi: 10.1016/0169-7439(87)80084-9
29	GRANATO D, SANTOS J S, ESCHER G B, et al. Use of principal component analysis (PCA) and hierarchical cluster analysis (HCA) for multivariate association between bioactive compounds and functional properties in foods: a critical perspective[J]. Trends in Food Science & Technology, 2018, 72: 83-90. DOI:10.1016/j.tifs.2017.12.006 doi: 10.1016/j.tifs.2017.12.006
30	QUAN W, WU Z L, JIAO Y, et al. Exploring the relationship between potato components and Maillard reaction derivative harmful products using multivariate statistical analysis[J]. Food Chemistry, 2021, 339: 127853. DOI:10.1016/j.foodchem. 2020.127853 doi: 10.1016/j.foodchem. 2020.127853
31	WOLPERT D H. Stacked generalization[J]. Neural Net- works, 1992, 5(2): 241-259. DOI:10.1016/s0893-6080(05)80023-1 doi: 10.1016/s0893-6080(05)80023-1

[1]	郑开友,任云,李洪雷,刘静,李强. 玉米/生姜套作模式下玉米株型与行宽对生姜光合特性及产量的影响[J]. 浙江大学学报（农业与生命科学版）, 2022, 48(3): 310-320.
[2]	郑福顺,王晓敏,李国花,李洪磊,周鹏泽,王林,白圣懿,刘珮君,张雪艳,胡新华,付金军,高艳明,李建设. 基于表型性状的宁夏番茄种质资源核心种质构建[J]. 浙江大学学报（农业与生命科学版）, 2021, 47(2): 171-181.
[3]	刘艺平,吴芳芳,贺丹,庄苑,孔德政. 基于花色表型的荷花品种数量分类[J]. 浙江大学学报（农业与生命科学版）, 2020, 46(3): 319-326.
[4]	赵燕,金俊杰,任敏敏,侯凤香,刘素贞,薛成俊,肖英平. 2种养殖模式下蛋鸭大肠埃希菌毒力基因和耐药特征分析[J]. 浙江大学学报（农业与生命科学版）, 2020, 46(2): 254-262.
[5]	许抗抗，丁天波，严毅，李灿，杨文佳. CO₂气调胁迫下烟草甲谷胱甘肽S-转移酶基因的表达分析[J]. 浙江大学学报(农业与生命科学版), 2017, 43(5): 599-607.
[6]	吕腾飞，周伟，孙永健，朱懿，严奉君，杨志远，马均. 不同秧龄和氮肥运筹对杂交籼稻株型的影响[J]. 浙江大学学报(农业与生命科学版), 2015, 41(2): 169-178.
[7]	杨志丽, 郭传龙, 刘蕾, 武孔焕, 王琳, 李昆志, 赵艳, 陈丽梅. 在烟草中过量表达大豆SGF14a增强转基因烟草对铝胁迫的耐受性[J]. 浙江大学学报(农业与生命科学版), 2015, 41(03): 285-292.
[8]	周佳萍，杨春元，吴春，王仁刚，史跃伟，谢升东，王志红，徐海明，任学良. 烟草核心种质库构建及遗传多样性研究(英文)[J]. 浙江大学学报(农业与生命科学版), 2014, 40(4): 440-450.
[9]	高玉龙1，桂毅杰2，肖炳光1,薄世平2，严广号2，樊龙江2. 烟草MITE位点间多态性 (IMP) 标记开发及其遗传作图应用[J]. 浙江大学学报(农业与生命科学版), 2012, 38(6): 655-661.
[10]	李伟成;王曙光;吴志庄;周妍;王树东;钟哲科. 2种大型丛生竹对氮输入的可塑性响应[J]. 浙江大学学报(农业与生命科学版), 2012, 38(5): 608-613.
[11]	许露露,潘文杰,张国平. 烟草自毒效应的品种间差异研究(英文)[J]. 浙江大学学报(农业与生命科学版), 2012, 38(2): 119-126.
[12]	程维舜,徐秋芳,黎飞,徐幼平,蔡新忠. 适于烟草脆裂病毒诱导的本氏烟基因沉默分析的对照载体构建(英文)[J]. 浙江大学学报(农业与生命科学版), 2012, 38(1): 10-20.
[13]	张建新,方依秋,丁彦芬,方炎明. 蕨类植物的叶绿素、光合参数与耐荫性[J]. 浙江大学学报(农业与生命科学版), 2011, 37(4): 413-420.
[14]	窦秉德, 侯北伟, 徐海明, 朱晓滨, 徐海风. 小麦雌性育性QTL的高效定位策略[J]. 浙江大学学报(农业与生命科学版), 2009, 35(6): 599-606.
[15]	姜磊　吴建祥　周雪平. 与烟草曲茎病毒相伴随的DNAβ的βC1基因原核表达及蛋白的单抗制备[J]. 浙江大学学报(农业与生命科学版), 2008, 34(2): 132-136.

Viewed

Full text

Abstract

Cited

Shared

Discussed