waterz
Pronounced water-zed. A simple watershed and region agglomeration library for affinity graphs.
Based on the watershed implementation of Aleksandar Zlateski and Chandan Singh.
Installation
Install c++ dependencies:
sudo apt install libboost-dev
Install from PyPI
pip install waterz
install from local version
pip install .
With conda on Windows:
git clone [email protected]:erjel/waterz.git
cd waterz
conda create -n waterz python boost-cpp
pip install .
Usage
import waterz
import numpy as np
# affinities is a [3,depth,height,width] numpy array of float32
affinities = ...
thresholds = [0, 100, 200]
segmentations = waterz.agglomerate(affinities, thresholds)
Development
Release to pypi
We use travis to create release
upgrade the version number in the setup.py
file, then
git tag v0.9.5
git push origin v0.9.5
the travis build system will get a TRAVIS_TAG
variable, and triger the twine upload
command.