Phenomenal

Phenomenal reconstructs and analyzes 3D shoot architecture from plant images to extract quantitative traits for plant phenotyping.

Key Features:

• Automatic Workflow System: Provides an end-to-end automatic workflow for data import, 3D reconstruction of plant architecture, and quantitative analysis, scalable to large datasets from high-throughput phenotyping platforms.
• Open Source Library: Exposes a flexible open-source library for development and comparison of algorithms for 3D shoot reconstruction.
• Integration Framework: Includes an integration framework to couple workflow outputs with existing models to support model-assisted phenotyping.
• Versatility Across Species and Data Formats: Handles datasets from diverse species, including dicotyledons and monocotyledons, and supports multiple imaging data formats and published datasets.
• Validation and Testing: Validated against manual measurements, synthetic data generated by 3D models, and other published datasets reproducing semi-automatic reconstruction workflows.

Scientific Applications:

• Shoot Architecture Quantification: Automates extraction of quantitative traits describing shoot topology and geometry for architectural studies.
• High-Throughput Phenotyping: Processes large-scale image datasets from high-throughput phenotyping platforms for trait extraction.
• Model-Assisted Phenotyping: Produces outputs suitable for coupling with computational models to enable model-assisted phenotyping analyses.
• Cross-Species Comparative Analysis: Enables comparative analyses across species, including dicotyledons and monocotyledons, using diverse datasets.

Methodology:

Performs automatic data import, algorithmic 3D reconstruction of shoot topology and geometry from images, and quantitative trait extraction within an integrated end-to-end workflow.