/cecog

A fast and cross-platform image analysis framework for fluorescence time-lapse microscopy and bioimage informatics.

Primary LanguagePythonOtherNOASSERTION

The CellCognition Project

Copyright (c) 2006 - 2018 Gerlich Lab, IMBA Vienna, Austria
The software is released under the terms of LGPL.

www.cellcognition-project.org
doc.cellcognition-project.org

⚠️ This project is orphaned and not actively developed anymore. Windows binaries might not work on Win10

Development build

The Extension depends on libtiff and libvigra. Adopt the library/include paths in the setup.cfg accordingly. On Unix (including OSX) like systems type:

Unixoides System

python setup.py build_ext --inplace

There's also am make file:

  • make clean
  • make inplace - install the C++-extenstion inplace ad builds *.rc and help files
  • make dmg - Binary installer for MacOSX

Windoze (7/8)

build_win64_bin.bat

System installation (e.g. on a cluster):

python setup.py install --prefix=<path-to-prefix>

Note: If you are using Windows an can not use the make fils, removethe build- and dist directories and also the file ./cecog/ccore/_cecog.pyd

Demo data (battery package)

The demo data contains:

  • a small set of raw images.
  • two classifiers for H2b and aTubulin.
  • a pre-configured settings file.

The demo data allows to try out image image segmentation, feature extraction, classifcation. To try ou also tracking, event selection and error correction, you need to download the full data set (206 frames, ~900Mb) from the cellcogition website