Related Experiment Video
Updated: Jan 11, 2026
08:25
Using Flatbed Scanners to Collect High-resolution Time-lapsed Images of the Arabidopsis Root Gravitropic Response
Published on: January 25, 2014
12.8K
Annotated 3D Point Cloud Dataset of Broad-Leaf Legumes Captured by High-Throughput Phenotyping Platform
Alexander Galba1, Jan Masner2, Jana Kholová1,3
1Department of Information Technologies, Faculty of Economics and Management, Czech University of Life Sciences, Kamýcká 129, Prague, 165 00, Czech Republic.
Scientific Data
|November 10, 2025
Summary
This study introduces annotated 3D plant scans of legumes, crucial for developing AI in plant phenomics. The dataset aids 3D computer vision model development for crop research.
Area of Science:
- Plant Phenomics
- Computer Vision
- Agricultural Science
Background:
- High-throughput phenotyping platforms are essential for modern crop research.
- Annotated 3D plant scan data is scarce, hindering AI model development.
- Legume crops are vital for global food security.
Purpose of the Study:
- To present a novel dataset of annotated 3D point cloud plant scans.
- To support the development of AI models for 3D computer vision in plant phenomics.
- To provide a resource for broad-leaf legume species research.
Main Methods:
- Utilized PlantEye(R) F600 technology for multispectral 3D scanning.
- Generated 223 scans of mungbean, common bean, cowpea, and lima bean.
- Performed organ-level segmentation and annotation for plant structures.
Main Results:
- A comprehensive dataset of annotated 3D plant scans was created.
- Detailed annotations include embryonic leaves, leaves, petioles, stems, and whole plants.
- The dataset includes preprocessing code and a MIAPPE-compliant data sheet.
Conclusions:
- The dataset addresses a critical need for annotated 3D plant data.
- Facilitates advancements in AI-driven plant phenotyping and crop improvement.
- The data and code are publicly available for further research and expansion.
