Horticultural Crop LiDAR Processing

This repository serves to store the code used for processing point clouds of horticultural crops. It was last updated on August 16, 2025.

File Locations

All plant point clouds and accompanying 2D images are stored in the Purdue University Research Repository "Horticultural Crop Point Clouds" dataset. The file setup of this repository mirrors that the user can simply copy this repository's files into a separate folder and add the point clouds, or add the point clouds to this repository in their correct location. Regardless, the point cloud files are much too large for GitHub to store.

Repository Structure

project_folders

• Biomass_and_Volume_Regressions - stores regressions for parameters and volume calculations.
• Garden_Plots - stores associated point clouds.
• Pepper_Variety_Trial - stores associated point clouds.
• Personal-Sized_Watermelon_Variety_Trial - stores associated point clouds.
• plot_figs - stores figures produced from analyses.
• Seedless_Watermelon_Variety_Trial - stores associated point clouds.
• Tomato_Fertilizer_Trial - stores associated point clouds.
• Transplant-Age_Plants - stores associated point clouds.

Environment

A description of the environment used for all processing and analysis is found in the environment.yml file in the main directory.

Important Scripts

1. bio_vol_reg_analysis.ipynb - analyzes the performance of the volume-structural characteristic regressions.
2. biomass_plots.ipynb - produces figures for physically measured structural characteristics.
3. descriptive_stats.ipynb - produces descriptive statistics for the point cloud dataset.
4. iou_stats.ipynb - produces descriptive statistics for the iou scores.
5. performance_test_pipeline.ipynb - calculates iou scores for processing evaluation.
6. processing_pipeline.ipynb - processes raw point clouds into finalized plant and non-plant point clouds.
7. sensor_validation_pipeline.ipynb - calculates difference between sensed and measured plant height.
8. volume_biomass_regressions_constrained_subset_2.ipynb - fits regression equations for parameters estimated from the calibration trial samples with volume metrics.
9. volume_pipeline.ipynb - calculates volume metrics for processed plant point clouds.