Automated and accurate segmentation of leaf venation networks via deep learning.

Abstract

Leaf vein network geometry can predict levels of resource transport, defence and mechanical support that operate at different spatial scales. However, it is challenging to quantify network architecture across scales due to the difficulties both in segmenting networks from images and in extracting multiscale statistics from subsequent network graph representations. Here we developed deep learning algorithms using convolutional neural networks (CNNs) to automatically segment leaf vein networks. Thirty-eight CNNs were trained on subsets of manually defined ground-truth regions from >700 leaves representing 50 southeast Asian plant families. Ensembles of six independently trained CNNs were used to segment networks from larger leaf regions (c. 100 mm 2 ). Segmented networks were analysed using hierarchical loop decomposition to extract a range of statistics describing scale transitions in vein and areole geometry. The CNN approach gave a precision-recall harmonic mean of 94.5% ± 6%, outperforming other current network extraction methods, and accurately described the widths, angles and connectivity of veins. Multiscale statistics then enabled the identification of previously undescribed variation in network architecture across species. We provide a LeafVeinCNN software package to enable multiscale quantification of leaf vein networks, facilitating the comparison across species and the exploration of the functional significance of different leaf vein architectures.

Plant phenotyping relevance

葉脈画像のセグメンテーションと形態統計抽出を深層学習で開発・検証し、再利用可能なソフトウェアとして提供しているため、植物フェノタイピング手法が中心である。

Code and data availability

The paper's Data and algorithm availability section explicitly deposits the LEAFVEINCNN GUI software with trained networks (Zenodo 4007731), the down-sampled CNN predictions, ground truths, and MATLAB analysis scripts (Zenodo 4008614), and all results (Zenodo 4008361), all openly available. These are paper-specific,公开,