Revealing Hierarchical Structure of Leaf Venations in Plant Science via Label-Efficient Segmentation: Dataset and Method

TL;DR

This paper introduces HALVS, the first public dataset for hierarchical leaf vein segmentation, and proposes a label-efficient learning method to improve segmentation with partial annotations, advancing plant science research.

Contribution

The paper provides a new hierarchical leaf vein dataset and a novel label-efficient segmentation approach leveraging partial labels, addressing key gaps in plant vein analysis.

Findings

HALVS dataset contains 5,057 high-resolution leaf images with detailed annotations.

The proposed method effectively utilizes partial labels for hierarchical vein segmentation.

Empirical results reveal new insights and challenges in leaf vein segmentation.

Abstract

Hierarchical leaf vein segmentation is a crucial but under-explored task in agricultural sciences, where analysis of the hierarchical structure of plant leaf venation can contribute to plant breeding. While current segmentation techniques rely on data-driven models, there is no publicly available dataset specifically designed for hierarchical leaf vein segmentation. To address this gap, we introduce the HierArchical Leaf Vein Segmentation (HALVS) dataset, the first public hierarchical leaf vein segmentation dataset. HALVS comprises 5,057 real-scanned high-resolution leaf images collected from three plant species: soybean, sweet cherry, and London planetree. It also includes human-annotated ground truth for three orders of leaf veins, with a total labeling effort of 83.8 person-days. Based on HALVS, we further develop a label-efficient learning paradigm that leverages partial label information, i.e. missing annotations for tertiary veins. Empirical studies are performed on HALVS, revealing new observations, challenges, and research directions on leaf vein segmentation.