Repeated Multiview Imaging for Estimating Seedling Tiller Counts of Wheat Genotypes Using Drones
作者机构：ETH ZurichInstitute of Agricultural SciencesUniversitätstrasse 28092 ZurichSwitzerland Delley Samen und Pflanzen AGRoute de Portalban 401567 DelleySwitzerland
出 版 物：《Plant Phenomics》 (植物表型组学（英文）)
年 卷 期：2020年第2卷第1期
页 面：279-298页
核心收录：
学科分类：0501[文学-中国语言文学] 0303[法学-社会学] 0710[理学-生物学] 0502[文学-外国语言文学] 0601[历史学-考古学] 1302[艺术学-音乐与舞蹈学] 1301[艺术学-艺术学理论] 09[农学] 0901[农学-作物学]
基 金：Innosuisse [KTI P-Nr 27059.2 PFLS-LS]
摘 要：Early generation breeding nurseries with thousands of genotypes in single-row plots are well suited to capitalize on high throughput ***,methods to monitor the intrinsically hard-to-phenotype early development of wheat are yet *** aimed to develop proxy measures for the rate of plant emergence,the number of tillers,and the beginning of stem elongation using drone-based *** used RGB images(ground sampling distance of 3mm pixel-1)acquired by repeated flights(≥2 flights per week)to quantify temporal changes of visible leaf *** exploit the information contained in the multitude of viewing angles within the RGB images,we processed them to multiview ground cover images showing plant pixel *** on these images,we trained a support vector machine for the beginning of stem elongation(GS30).Using the GS30 as key point,we subsequently extracted plant and tiller counts using a watershed algorithm and growth modeling,*** results show that determination coefficients of predictions are moderate for plant count(R^(2)=0:52),but strong for tiller count(R^(2)=0:86)and GS30(R^(2)=0:77).Heritabilities are superior to manual measurements for plant count and tiller count,but inferior for GS30 *** the selection intensity due to throughput may overcome this *** image traits can replace hand measurements with high efficiency(85-223%).We therefore conclude that multiview images have a high potential to become a standard tool in plant phenomics.
维普期刊数据库