Skip to content

Rdataflow/pyFINT

 
 

Repository files navigation

pyFINT: Python implementation of FINT

Overview

pyFINT is a Python implementation of the FINT (Find INdividual Trees) tool from ecorisQ for detecting individual trees in a normalized crown height model. The program has been created as part of the swiss FINT-CH research project.

TL;DR

docker build --rm -t pyfint . docker run --rm -v $(pwd)/output:/pyfint/output -ti pyfint

Usage

The program has been written for Python 3.x. See requirements.txt for dependencies. The classes of the python version were kept close to the original C++ implementation. Their use is therefore not necessarily very pythonic. Details about the FINT algorithm and the program outputs can be found in the original tool's user manual: https://www.ecorisq.org/docs/FINT_manual_EN.pdf. Compared to the original version, pyFINT only supports normalized surface models in the GeoTIFF or Esri ASCII format as input and it doesn't support the calculation of a normalized surface model form a surface and terrain model. Additionally, pyFINT supports optional resizing and gauss filtering of the surface model, not yet available in the original C++ program. The example_main.py shows examples for the most common use cases of pyFINT.

Inputs:

  • Normalized surface model as GeoTIFF or ASCII
  • Optional: digital terrain model as GeoTIFF or ASCII Outputs:
  • *.TXT and *.CSV files with the detected individual trees
  • schema.ini with an explanation of the columns in the *.TXT and *.CSV
  • Optional: If a resizing or filtering option is configured, the modified normalized surface model is output as a GeoTIFF

Links

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 99.9%
  • Dockerfile 0.1%