摘要
摘要 株高是春燕麦（Avena sativa L.）的关键经济性状之一，也是倒伏的主要决定因素，进而决定产量潜力。理解株高变异的遗传基础可显著有益于植物改良工作。尽管株高易于测量，但在大量小区和环境中进行测量时可能十分费力。利用安装在无人航空系统（UAS）上的传感器所获取的航空图像快速且经济高效地估算株高，可作为加速育种计划目标的替代方法。为实现这一目标，利用数字表面模型在小区水平上对一个全球燕麦种质材料群体进行了株高自动估算评估。植被指数由UAS RGB（红、绿、蓝）色彩和在不同间隔从四个地点采集的多光谱影像计算得出。将植被指数与常规测量进行建模以评估准确性。利用UAS和地面真值数据进行全基因组关联作图，以全面理解株高的遗传结构。最后，我们讨论了针对UAS和地面真值数据的基因组预测，以将UAS在实际植物育种中的使用置于具体背景中。
Abstract Plant height is one of the key economic traits in spring oat ( Avena sativa L.) and a major determinant of lodging, and consequently, yield potential. Understanding the genetic basis of the variation in plant height can significantly benefit plant improvement efforts. Although plant height is straightforward to measure, it can be intense when conducted across numerous plots and environments. A rapid and cost‐effective estimation of plant height from aerial images obtained from sensors mounted onto uncrewed aerial systems (UAS) can be an alternative approach to accelerate the goals of breeding programs. To achieve this, a global panel of oat accessions was evaluated for automatic plant height estimation at the plot‐level using digital surface models. Vegetation indices were calculated from the UAS RGB (red, green, blue) color and multispectral imagery collected from four sites at different intervals. Vegetation indices were modeled with conventional measurements to assess accuracy. Genome‐wide association mapping was performed using UAS and ground truth data to gain a comprehensive understanding of the genetic architecture of plant height. Finally, we discuss genomic prediction for both UAS and ground truth data to contextualize the use of UAS in practical plant breeding.