ReLeaf: Benchmarking Leaf Segmentation across Domains and Species

2026-05-05

AI Summary

Leaf-level instance segmentation in precision agriculture faces challenges due to limited species coverage and a lack of systematic evaluation. This work addresses these limitations by integrating four public datasets to establish a new cross-species, cross-scenario leaf segmentation benchmark encompassing 23 crop and weed species. The authors further enhance the CropAndWeed dataset through a semi-automated annotation pipeline. A comprehensive evaluation of single-stage, two-stage, and Transformer-based instance segmentation models identifies YOLO26 as the top-performing architecture. Employing a joint training strategy, the model achieves 83.9% mAP50-95 on the original test set and 40.2% on the new benchmark—significantly outperforming models trained solely on laboratory data—and demonstrates markedly improved cross-domain generalization capability.

Abstract

Rising global food demand and growing climate pressure increase the need for sustainable, precise agricultural practices. Automated, individualized plant treatment relies on fine-grained visual analysis, yet leaf-level segmentation remains underexplored despite its value for assessing crop health, growth dynamics, yield potential and localized stress symptoms. Progress is limited by a lack of dedicated datasets, especially regarding species coverage, and by the absence of systematic evaluations of modern instance-segmentation architectures for this task. We address these gaps by surveying current data and identifying four suitable, publicly available leaf-segmentation datasets. Using them, we compare one-stage, two-stage and Transformer-based detectors and identify a YOLO26 model configuration to provide the best trade-off for real-world precision-agriculture tasks. Extensive cross-domain generalization experiments reveal substantial performance drops across plant species and recording setups, especially for models trained solely on laboratory data. To strengthen data availability, we introduce a new benchmark dataset with leaf-level masks for 23 plant species, created via semi-automatic annotation of selected CropAndWeed images. A model trained on all four existing datasets achieves a mean mAP50-95 of 83.9% across their corresponding test sets and 40.2% on our new benchmark, demonstrating improved generalization and highlighting the need for diverse leaf-segmentation datasets in robust precision agriculture.

Problem

• leaf segmentation
• instance segmentation
• precision agriculture
• cross-domain generalization
• plant species

Innovation

• leaf segmentation
• instance segmentation
• cross-domain generalization
• precision agriculture
• benchmark dataset