A desktop application for digitising and managing agricultural field trial plot boundaries over drone orthomosaic imagery. Built with Streamlit and Python geospatial libraries, it runs locally in your web browser with no cloud dependency.
Developed at the Department of Primary Industries and Regional Development (DPIRD), Western Australia, as part of the Australian Plant Phenomics Network (APPN).
- Generate Grid — Create regular plot grids over drone orthomosaics by drawing a trial boundary and specifying banks, rows, buffer, and plot dimensions.
- Edit Grid — Interactive browser-based polygon editor with drag, vertex editing, multi-select, copy/paste, measurements, undo, and keyboard shortcuts.
- Convert File — Convert between Shapefile, GeoJSON, GeoPackage, and KML formats with optional CRS reprojection (GDA2020, GDA94, WGS 84, or custom EPSG).
- Cropping Tool — Crop rasters (orthophotos, DSMs, GeoTIFFs) to individual plot polygon boundaries, producing one file per plot.
All data is processed locally. No internet connection is required after installation.
- Download and extract
DPIRD_Field_Mapping_Tool_windows_v1.6.0. - Place
Miniconda3-latest-Windows-x86_64.exein the folder. - Double-click
Install_DPIRD_Tool.batand pressY. - Double-click
Run_DPIRD_Tool.batto launch.
- Download and extract
DPIRD_Field_Mapping_Tool_linux_v1.6.0. - Ensure Anaconda or Miniconda is installed.
- Run:
chmod +x install_dpird_tool.sh run_dpird_tool.sh ./install_dpird_tool.sh ./run_dpird_tool.sh
The tool opens in your browser at http://localhost:8501.
├── DPIRD_Field_Mapping_Tool_windows_v1.6.0/ # Windows distribution
│ ├── app/
│ │ └── field_mapping_tool.py
│ ├── bin/
│ │ └── logo.png
│ ├── Install_DPIRD_Tool.bat
│ ├── Run_DPIRD_Tool.bat
│ └── readme.md
│
├── DPIRD_Field_Mapping_Tool_linux_v1.6.0/ # Linux / macOS distribution
│ ├── app/
│ │ └── field_mapping_tool.py
│ ├── bin/
│ │ └── logo.png
│ ├── install_dpird_tool.sh
│ ├── run_dpird_tool.sh
│ └── readme.md
│
├── LICENSE
└── README.md
The application code (field_mapping_tool.py) is identical across both distributions. The platform folders differ only in their installation and launch scripts.
| Windows | Linux / macOS | |
|---|---|---|
| OS | Windows 10+ (64-bit) | Ubuntu 20.04+, Fedora, macOS 11+ |
| Python | Bundled via Miniconda | User-installed Anaconda/Miniconda |
| Disk space | ~2 GB | ~2 GB |
| Internet | First-time setup only | First-time setup only |
Installed automatically by the setup scripts:
conda-forge: numpy, pandas, geopandas, rasterio, fiona, shapely, pyproj, pillow, matplotlib, psutil
pip: streamlit, streamlit-folium, folium, leafmap, localtileserver, scooby
DPIRD_Field_Mapping_Tool_Documentation.pdf provides the full usage instructions for the tool.
This project is licensed under the MIT License — see LICENSE for details.
Bipul Neupane, PhD Research Scientist, DPIRD Node Australian Plant Phenomics Network (APPN) Department of Primary Industries and Regional Development (DPIRD), Western Australia bipul.neupane@dpird.wa.gov.au
Bipul Neupane: Github
The plot generation logic and parameters are inspired by FieldImageR and PlimanShiny Tools. The GUI was developed with the assistance of Claude AI by Anthropic.