Training instance segmentation neural network with synthetic datasets for crop seed phenotyping

Abstract

In order to train the neural network for plant phenotyping, a sufficient amount of training data must be prepared, which requires time-consuming manual data annotation process that often becomes the limiting step. Here, we show that an instance segmentation neural network aimed to phenotype the barley seed morphology of various cultivars, can be sufficiently trained purely by a synthetically generated dataset. Our attempt is based on the concept of domain randomization, where a large amount of image is generated by randomly orienting the seed object to a virtual canvas. The trained model showed 96% recall and 95% average Precision against the real-world test dataset. We show that our approach is effective also for various crops including rice, lettuce, oat, and wheat. Constructing and utilizing such synthetic data can be a powerful method to alleviate human labor costs for deploying deep learning-based analysis in the agricultural domain.

Plant phenotyping relevance

合成データとインスタンスセグメンテーションによる種子形態フェノタイピング手法を開発し、実画像で性能検証しているため、方法が研究の中心である。

Code and data availability

The authors publicly release both the synthetic and real-world seed image datasets and the analysis code (Mask R-CNN deployment and multivariate analysis notebooks) via their GitHub repository, explicitly stated in Data availability and Code availability sections.