High-Throughput Phenotyping of Individual Wheat Stem and Ear Traits with Machine Vision
DING QiShuo1, LI HaiKang1, SUN KeRun2, HE RuiYin1, WANG XiaoChan1, LIU FuXi1, LI Xiang11 College of Engineering, Nanjing Agricultural University/Key Laboratory of Intelligent Agricultural Equipment of Jiangsu Province, Nanjing 210031 2 Jiangsu Yinhua Chunxiang Machinery Manufacturing Co. Ltd., Lianyungang 222200, Jiangsu
Abstract 【Objective】 High-throughput phenotyping (HTP) is not only an important tool of modern agriculture for crop breeding, but also a powerful means to illustrate physiological and ecological mechanisms of crops in the field. However, the basic features of structural components of each HTP tools have to be illustrated. It is therefore necessary to investigate what a technical feature is applicable to machine vision based HTP system.【Method】 An image-processing tool was developed to measure stem-and-ear level traits of each individual wheat stem. Three wheat species, i.e. Ningmai 13, Luyuan 502 and Zhengmai 9023, were used for plot experiment analysis. The wheat was sown with luffer board having equally spaced seeding holes. The precision seeding tools were applied to control wheat population accurately. At the maturity of post-paddy wheat, the integrated image of stem and ear was obtained, and the image was subjected to gray enhancement, histogram equalization, S component extraction, Otsu threshold segmentation, stem and ear separation, and stem and ear morphology parameters. The morphological parameters of the individual organs per stem-panicle of the extracted post-paddy wheat included stem length, average stem width, stem projection area, stem circumference, ear length, average ear width, ear projection area and ear circumference. In addition, traditional methods of measurement were used to derive single leaf weight, single stem weight, single ear weight and single ear yield etc. Linear, quadratic, extended and exponential models were applied for the regression on the collected multi-dimensional data sets, including correlations between ear and stem level biomass and individual ear grain yield, interrelationships among morphological parameters of stem and ear and single ear grain yield. Correlation analysis and regression analysis were performed on the processed indices of wheat. Based on this case study, some key aspects of technologies were discussed concerning on the application of machine vision tools on high-throughput phenotyping in the field.【Result】Results showed that correlation coefficients of individual stem and leaf weight with individual ear grain yield decreased steadily from Ningmai 13 to Luyuan 502, and till Zhengmai 9023. Correlation coefficient of stem and ear morphological parameters with individual ear grain yield was significantly lower than that among the biomasses. However, composite morphological parameter, which integrated single ear projection area and single ear length, was found significantly correlated with individual ear grain yield. The best regression model for the correlation between stem and ear biomass and individual ear grain yield of the three wheat species were different. Morphological parameters derived from ear images failed to predict individual ear grain yield precisely. However, combined morphological parameters from wheat stem and wheat ear revealed the best result of regression with extension models. Composite morphological stem-and-ear level traits of individual wheat stem provided more accurate prediction on the ear-derived grain yield, which could make the yield prediction with growth-stage traits collected with machine vision technically possible. Machine vision tools of HTP provided a much higher sets of agronomic trait indices as compared with traditional methods, providing more options for the illustration on the correlations among agronomic traits and path-analysis on crop yield. It in turn resulted into high-dimensional data sets and technical difficulties impeding the identification on valuable information. 【Conclusion】A basic infrastructure of HTP machine vision tools for field wheat stand was defined as multi-scale and automatic adaptation aspect. It should be autonomously adaptable to multi-scales concerning with the field, crop stand, individual crop and organ-level traits of each individual crop. It also provided traits identification and calculation with statistical analysis on different physiological periods of wheat, e.g. seedling stage, tillering stage, jointing stage etc. Meanwhile, in each development stage of the machine-vision HTP tools and for each functional module, in-depth involvement of agronomical calibration was required. In safeguarding the reliability of machine-vision tools, standardization on referencing HTP-derived traits was also necessary. Keywords:machine vision;individual stem and ear;high-throughput analysis;trait indices;ear-level grain yield
PDF (1169KB)元数据多维度评价相关文章导出EndNote|Ris|Bibtex收藏本文 本文引用格式 丁启朔, 李海康, 孙克润, 何瑞银, 汪小旵, 刘富玺, 厉翔. 基于机器视觉的稻茬麦单茎穗高通量表型分析[J]. 中国农业科学, 2020, 53(1): 42-54 doi:10.3864/j.issn.0578-1752.2020.01.004 DING QiShuo, LI HaiKang, SUN KeRun, HE RuiYin, WANG XiaoChan, LIU FuXi, LI Xiang. High-Throughput Phenotyping of Individual Wheat Stem and Ear Traits with Machine Vision[J]. Scientia Acricultura Sinica, 2020, 53(1): 42-54 doi:10.3864/j.issn.0578-1752.2020.01.004
SEY:单穗籽粒产量;SLW:单叶片质量;SSW:单茎秆质量;SEW:单穗质量。下同 SEY: Single ear yield; SLW: Single leaf weight; SSW: Single stem weight; SEW: Single ear weight. The same as below
Table 2 表2 表2小麦单茎穗茎秆和麦穗形态参数与单穗籽粒产量回归模型 Table 2The regression model between stem & ear morphological parameters per stem-panicle and ear-derived grain yield of wheat
拟合模型 Fitting model
模型类型 Model type
模型方程 Model equation
麦穗形态参数与单穗籽粒产量 Ear morphological parameters and ear-derived grain yield
SEL:单穗长;SEAW:单穗平均宽度;SEA:单穗投影面积;SEC:单穗周长;SSL:单茎长;SSAW:单茎平均宽度;SSA:单茎投影面积;SSC:单茎周长。下同 SEL: Single ear length; SEAW: Average width of single ear; SEA: Single ear area; SEC: Single ear circumference; SSL: Single stem length; SSAW: Average width of single stem; SSA: Single stem area; SSC: Single stem circumference. The same as below
Table 4 表4 表4小麦单茎穗茎秆和麦穗形态参数与单穗籽粒产量的相关性 Table 4Correlations between stem and ear morphological parameters per stem-panicle and ear-derived grain yield of wheat
PANY, XIONG QX, ZHANG WY . Research advances in digital image technology and its application in crop phenotype Journal of Yangtze University(Natural Science Edition), 2016,13(21):38-41, 46. (in Chinese) [本文引用: 1]
SINGHA, GANAPATHYSUBRAMANIANB, SINGH AK, SARKARS . Machine learning for high-throughput stress phenotyping in plants , 2015,21(2):110-124. [本文引用: 1]
GOGGIN FL, LORENCEA, TOPP CN . Applying high-through-put phenotyping to plant-insect interactions: Picturing more resistant crops , 2015,9:69-76. [本文引用: 1]
FAN MY, MAQ, LIU JM, WANGQ, WANGY, DUAN XC . Counting method of wheatear in field based on machine vision technology Transactions of the Chinese Society for Agricultural Machinery, 2015,46(S1):234-239. (in Chinese) [本文引用: 1]
LU WC, LUOB, PAN DY, ZHAOY, YU CH, WANGC . Synchronous measurement of wheat ear length and spikelets number based on image processing Journal of Chinese Agricultural Mechanization, 2016,37(6):210-215. (in Chinese) [本文引用: 1]
LIUT, SUN CM, WANG LJ, ZHONG XC, ZHU XK, GUO WS . In-field wheatear counting based on image processing technology Transactions of the Chinese Society for Agricultural Machinery, 2014,45(2):282-290. (in Chinese) [本文引用: 1]
DU SW, LI YN, YAOM, LIL, DING QS, HE RY . Counting method of grain number based on wheatear spikelet image segmentation Journal of Nanjing Agricultural University, 2018,41(4):742-751. (in Chinese) [本文引用: 3]
ZHAO SQ, LI YN, DING WM, Lü JY, WANG XY . Relative analysis between image characteristics of panicle structure and spikelet number Transactions of the Chinese Society for Agricultural Machinery, 2014,45(12):323-328. (in Chinese) [本文引用: 1]
HUP, CHAPMAN SC, WANGX, POTGIETERA, DUANT, JORDAND, GUOY, ZHENGB . Estimation of plant height using a high throughput phenotyping platform based on unmanned aerial vehicle and self-calibration: Example for sorghum breeding , 2018,95:24-32. [本文引用: 1]
NIU QL, FENG HK, YANG GJ, LI CC, YANGH, XUB, ZHAO YX . Monitoring plant height and leaf area index of maize breeding material based on UAV digital images Transactions of the Chinese Society of Agricultural Engineering, 2018,34(5):73-82. (in Chinese) [本文引用: 1]
FANC, XIAX, SHI XF, HOU LL . Research on wheat variety classification based on optimized artificial neural network Journal of Henan University of Technology(Natural Science Edition), 2012,33(4):72-76. (in Chinese) [本文引用: 1]
HE SM, LI ZL, HE ZH . Classification of wheat cultivar by digital image analysis Scientia Agricultura Sinica, 2005,38(9):1869-1875. (in Chinese) [本文引用: 1]
MEBATSION HK, PALIWALJ, JAYAS DS . A novel, invariant elliptic Fourier coefficient based classification of cereal grains , 2012,111(4):422-428. [本文引用: 1]
TAHIR AR, NEETHIRAJANS, JAYAS DS, SHAHIN MA, SYMONS SJ, WHITE N DG . Evaluation of the effect of moisture content on cereal grains by digital image analysis , 2007,40(9):1140-1145. [本文引用: 1]
KIPPS, MISTELEB, BARESELP, SCHMIDHALTERU . High-throughput phenotyping early plant vigour of winter wheat , 2014,52:271-278. [本文引用: 1]
ZHANGM, SUNH, LI MZ, ZHANGQ, ZHENG LH . Prediction of winter wheat tiller number based on 4-waveband crop monitor with spectral reflectance Transactions of the Chinese Society for Agricultural Machinery, 2016,47(9):341-347. (in Chinese) [本文引用: 1]
REN JQ, WU SR, LIUB, CHEN ZX, LIU XR, LIH . Retrieving winter wheat above-ground dry biomass based on hyperion hyperspectral imagery Transactions of the Chinese Society for Agricultural Machinery, 2018,49(4):199-211. (in Chinese) [本文引用: 1]
WUD, YANG WN, NIU ZY, HUANG CL . Non-destructive detection of wheat tiller morphological traits based on X-ray CT technology Transactions of the Chinese Society of Agricultural Engineering, 2017,33(14):196-201. (in Chinese) [本文引用: 2]
YANGW, XUX, CHENS, DUANL, LUOQ, CHENS, ZENGS, LIUQ . High-throughput measurement of rice tillers using a conveyor equipped with x-ray computed tomography , 2011,82(2):025102-025109. [本文引用: 1]
NEETHIRAJANS, JAYAS DS, WHITE N DG . Detection of sprouted wheat kernels using soft X-ray image analysis , 2007,81(3):509-513. [本文引用: 1]
KRONENBERGL, YUK, WALTERA, HUNDA . Monitoring the dynamics of wheat stem elongation: Genotypes differ at critical stages , 2017,213(7):157-170. [本文引用: 1]
REISERD, VáZQUEZ-ARELLANOM, PARAFOROS DS, MARRIDO-IZARDM, GRIEPENTROG HW . Iterative individual plant clustering in maize with assembled 2D LiDAR data , 2018,99:42-52. [本文引用: 1]
JIMENEZ-BERNI JA, DEERY DM, ROZAS-LARRAONDOP, CONDON AG, REBETZKE GJ, GREGJ, JAMES RA, BOVILL WD, FURBANK RT, SIRAULT X RR . High throughput determination of plant height, ground cover, and above-ground biomass in wheat with LiDAR , 2018,9(237):1-18. [本文引用: 1]
ZHOUJ, TARDIEUF, PRIDMORET, DOONANJ, REYNOLDSD, HALLN, GRIFFITHSS, CHENGT, ZHUY, WANG XE, JIANGD, DING YF . Plant phenomics: History, present status and challenges Journal of Nanjing Agricultural University, 2018,41(4):580-588. (in Chinese) [本文引用: 3]
JINX, LIUS, BARETF, HEMERLéM, COMARA . Estimates of plant density of wheat crops at emergence from very low altitude UAV imagery , 2017,198:105-114. [本文引用: 1]
KIMD, YUN HS, JEONGS, KWONY, KIMS, LEE WS, KIMH . Modeling and testing of growth status for Chinese cabbage and white radish with UAV-Based RGB imagery , 2018,10(4):563-588. [本文引用: 1]
KHANZ, RAHIMI-EICHIV, HAEFELES, GARNETTT, MIKLAVCIC SJ . Estimation of vegetation indices for high-throughput phenotyping of wheat using aerial imaging , 2018,14(1):20. [本文引用: 1]
HE JQ, HARRISON RJ, LIB . A novel 3D imaging system for strawberry phenotyping , 2017,13(1):93-101. [本文引用: 1]
WALTERJ, EDWARDSJ, MCDONALDG, KUCHELH . Photogrammetry for the estimation of wheat biomass and harvest index , 2018,216:165-174. [本文引用: 1]
CHEN YL, YANGP, ZHANG WY, ZHANG WX, ZHU YP, LI SJ, GONG FJ, BI HB, YUET, CAO HX . Study on main geometrical parameter models of leaf blade at pre-overwintering of winter wheat Journal of Triticeae Crops, 2015,35(12):1719-1726. (in Chinese) [本文引用: 2]
CHEN LG, LIU HJ, SHEN MX, ZHENG JC . Effects of different seeding modes on grain yield formation of wheat Jiangsu Journal of Agricultural Sciences, 2015,31(4):786-791. (in Chinese) [本文引用: 1]
ZHANG MW, MAQ, DING JF, LI CY, ZHU XK, FENG CN, GUO WS . Effect of density and nitrogen application on clum lodging resistance and yield of late sowing wheat Journal of Triticeae Crops, 2018,38(5):584-592. (in Chinese) [本文引用: 1]
REUZEAUC, PENJ, FRANKARDV, WOLFJ, PEERBOLTER, BROEKAERTW, CAMPW . TraitMill: A discovery engine for identifying yield-enhancement genes in cereals , 2005,3(5):753-759. [本文引用: 1]
WUS, GUO XY, MIAOT, XIAO BX, WEN WL, ZHAO CJ . Construction and application of intelligent plant platform based on Multi-Agent Journal of Agricultural Science and Technology, 2017,19(5):60-69. (in Chinese) [本文引用: 2]
FENGH, XIONG LZ, CHEN GX, YANG WN, LIUQ . The segmentation of leaf and stem of individual rice plant with hyperspectral imaging system and principal component analysis Acta Laser Biology Sinica, 2015,24(1):31-37. (in Chinese) [本文引用: 1]
OTSUN . A threshold selection method from gray-level histograms , 1979,9(1):62-66. [本文引用: 1]
CHEN YL, YANGP, ZHANG WY, ZHANG WX, ZHU YP, LI SJ, GONG FJ, BI HB, YUET, CAO HX . Biomass-based main spike morphological parameter model for winter wheat Acta Agronomica Sinica, 2017,43(3):399-406. (in Chinese) [本文引用: 2]
QIZ, ZHAO GC, CHANG XH, WANG DM, TAO ZQ, YANG YS, WANGM, FAN ZQ, GUO MM, WANGY, LIUT, LIU XC . Correlation and path analysis on wheat yield and agronomic indices Crops, 2016(3):45-50. (in Chinese) [本文引用: 1]
ZHOU YH, ZHU XK, GUO WS, FENG CN . Analysis of yield and yield components of wheat after rice on medium-high-yielding level Journal of Triticeae Crops, 2018,28(3):293-297. (in Chinese) [本文引用: 1]
GONG HJ, JI CY . Estimating yield of wheat spike with its texture features based on image processing technology Transactions of the Chinese Society for Agricultural Machinery, 2007,38(12):116-119. (in Chinese) [本文引用: 1]
QIU RC, MIAO YL, ZHANGM, LIH, SONGH . Modeling and verification of maize biomass based on linear regression analysis Transactions of the Chinese Society of Agricultural Engineering, 2018,34(10):131-137. (in Chinese) [本文引用: 2]
ZHAO CJ, LU SL, GUO XY, DU JJ, WEN WL, MIAOT . Advances in research of digital plant: 3D digitization of plant morphological structure Scientia Agricultura Sinica, 2015,48(17):3415-3428. (in Chinese) [本文引用: 2]
SHAO QQ, ZHOUQ, WANGX, CAIJ, HUANGM, DAI TB, JIANGD . Effects of planting density on stem morphological characteristics, chemical composition and lodging resistance of different wheat varieties Journal of Nanjing Agricultural University, 2018,41(5):808-816. (in Chinese) [本文引用: 1]
LEI XJ, TANGL, ZHANG YH, JIANG HY, CAO WX, ZHUY . Geometric model and visualization of wheat spike Transactions of the Chinese Society of Agricultural Engineering, 2011,27(3):179-184. (in Chinese) [本文引用: 1]
YANGY, LIUB, LIU XJ, LIU LL, FAN XM, CAO WX, ZHUY . Comparison of phasic development models in wheat Journal of Nanjing Agricultural University, 2014,37(1):6-14. (in Chinese) [本文引用: 1]