StomataSeg: Semi-Supervised Instance Segmentation for Sorghum Stomatal Components

TL;DR

This paper introduces a semi-supervised instance segmentation framework for sorghum stomata, leveraging patch-based preprocessing and pseudo-labeling to improve detection of tiny structures, enabling scalable phenotyping for climate-resilient agriculture.

Contribution

It presents a novel semi-supervised segmentation method tailored for small stomatal components, with a large annotated dataset and significant performance improvements over existing models.

Findings

Semantic mIoU increased from 65.93% to 70.35%.

Instance AP improved from 28.30% to 46.10%.

Patch-based semi-supervised learning enhances tiny structure segmentation.

Abstract

Sorghum is a globally important cereal grown widely in water-limited and stress-prone regions. Its strong drought tolerance makes it a priority crop for climate-resilient agriculture. Improving water-use efficiency in sorghum requires precise characterisation of stomatal traits, as stomata control of gas exchange, transpiration and photosynthesis have a major influence on crop performance. Automated analysis of sorghum stomata is difficult because the stomata are small (often less than 40 $\mu$m in length in grasses such as sorghum) and vary in shape across genotypes and leaf surfaces. Automated segmentation contributes to high-throughput stomatal phenotyping, yet current methods still face challenges related to nested small structures and annotation bottlenecks. In this paper, we propose a semi-supervised instance segmentation framework tailored for analysis of sorghum stomatal components. We collect and annotate a sorghum leaf imagery dataset containing 11,060 human-annotated patches, covering the three stomatal components (pore, guard cell and complex area) across multiple genotypes and leaf surfaces. To improve the detection of tiny structures, we split high-resolution microscopy images into overlapping small patches. We then apply a pseudo-labelling strategy to unannotated images, producing an additional 56,428 pseudo-labelled patches. Benchmarking across semantic and instance segmentation models shows substantial performance gains: for semantic models the top mIoU increases from 65.93% to 70.35%, whereas for instance models the top AP rises from 28.30% to 46.10%. These results demonstrate that combining patch-based preprocessing with semi-supervised learning significantly improves the segmentation of fine stomatal structures. The proposed framework supports scalable extraction of stomatal traits and facilitates broader adoption of AI-driven phenotyping in crop science.