A deep learning-based approach for high-throughput hypocotyl phenotyping

Abstract

Hypocotyl length determination is a widely used method to phenotype young seedlings. The measurement itself has been developed from using rulers and millimeter papers to the assessment of digitized images, yet it remained a labour-intensive, monotonous and time consuming procedure. To make high-throughput plant phenotyping possible, we developed a deep learning-based approach to simplify and accelerate this method. Our pipeline does not require a specialized imaging system but works well with low quality images, produced with a simple flatbed scanner or a smartphone camera. Moreover, it is easily adaptable for a diverse range of datasets, not restricted to Arabidopsis thaliana . Furthermore, we show that the accuracy of the method reaches human performance. We not only provide the full code at https://github.com/biomag-lab/hypocotyl-UNet , but also give detailed instructions on how the algorithm can be trained with custom data, tailoring it for the requirements and imaging setup of the user. One-sentence summary A deep learning-based algorithm, providing an adaptable tool for determining hypocotyl or coleoptile length of different plant species.

Plant phenotyping relevance

幼植物の表現型である胚軸・子葉鞘長を画像から高スループットに推定する深層学習手法を開発しており、表現型取得・抽出法が研究の中心である。

Code and data availability

The paper explicitly states that the full open-source analysis code (U-Net-based hypocotyl segmentation/measurement pipeline) is available on GitHub and that the training images (annotated Arabidopsis, Sinapis, Brachypodium seedling images) are publicly available on Kaggle. Both are paper-specific, public, and directly