Lagergren John, Pavicic Mirko, Chhetri Hari B, York Larry M, Hyatt Doug, Kainer David, Rutter Erica M, Flores Kevin, Bailey-Bale Jack, Klein Marie, Taylor Gail, Jacobson Daniel, Streich Jared
Biosciences Division, Oak Ridge National Laboratory, Oak Ridge, TN, USA.
Department of Applied Mathematics, University of California, Merced, CA, USA.
Plant Phenomics. 2023 Jul 28;5:0072. doi: 10.34133/plantphenomics.0072. eCollection 2023.
Plant phenotyping is typically a time-consuming and expensive endeavor, requiring large groups of researchers to meticulously measure biologically relevant plant traits, and is the main bottleneck in understanding plant adaptation and the genetic architecture underlying complex traits at population scale. In this work, we address these challenges by leveraging few-shot learning with convolutional neural networks to segment the leaf body and visible venation of 2,906 leaf images obtained in the field. In contrast to previous methods, our approach (a) does not require experimental or image preprocessing, (b) uses the raw RGB images at full resolution, and (c) requires very few samples for training (e.g., just 8 images for vein segmentation). Traits relating to leaf morphology and vein topology are extracted from the resulting segmentations using traditional open-source image-processing tools, validated using real-world physical measurements, and used to conduct a genome-wide association study to identify genes controlling the traits. In this way, the current work is designed to provide the plant phenotyping community with (a) methods for fast and accurate image-based feature extraction that require minimal training data and (b) a new population-scale dataset, including 68 different leaf phenotypes, for domain scientists and machine learning researchers. All of the few-shot learning code, data, and results are made publicly available.
植物表型分析通常是一项耗时且昂贵的工作，需要大批研究人员精心测量与植物生物学相关的性状，并且是在种群规模上理解植物适应性以及复杂性状背后的遗传结构的主要瓶颈。在这项工作中，我们通过利用卷积神经网络的少样本学习来分割在田间获取的2906张叶片图像的叶片主体和可见叶脉，从而应对这些挑战。与先前的方法相比，我们的方法具有以下特点：（a）不需要实验或图像预处理；（b）使用全分辨率的原始RGB图像；（c）训练所需的样本非常少（例如，叶脉分割仅需8张图像）。使用传统的开源图像处理工具从所得分割结果中提取与叶片形态和叶脉拓扑相关的性状，通过实际物理测量进行验证，并用于进行全基因组关联研究以鉴定控制这些性状的基因。通过这种方式，当前工作旨在为植物表型分析领域提供：（a）需要最少训练数据的快速准确的基于图像的特征提取方法；（b）一个新的种群规模数据集，包括68种不同的叶片表型，供领域科学家和机器学习研究人员使用。所有少样本学习代码、数据和结果均已公开。
