Plant 3D

Plant 3D performs segmentation and analysis of 3D plant point clouds to enable plant phenotyping, biomass estimation, and morphological and genotype-phenotype studies.

Key Features:

• Core Phenotyping Tasks: Performs classification of point clouds into stem and lamina, graph skeletonization of stem points, segmentation of individual laminae, and labeling of entire leaves.
• High-Performance Rendering Engine: Provides a fast 3D rendering engine capable of visualizing plant point clouds composed of millions of points.
• Advanced Analytical Algorithms: Implements graph-theoretic and machine-learning algorithms for analysis of plant architecture and trait quantification.
• Data Compatibility and Visualization: Imports 3D point clouds (pcd or txt) and mesh files (obj) and supports visualization of these formats.

Scientific Applications:

• Biomass Quantification: Facilitates estimation of plant biomass through structural classification and analysis of 3D point clouds.
• Morphological Analysis: Enables segmentation and labeling for detailed morphological studies of plant form and structure.
• Genotype-Phenotype Studies: Supports linking phenotypic 3D structural data to genetic information for genotype-phenotype investigations.

Methodology:

Implemented in C++ and using OpenGL, QT, TensorFlow, and the Point Cloud Library (PCL), with graph-theoretic and machine-learning algorithms for point cloud processing.