This is an alpha version of the Max-tree toolbox. The toolbox was implemented in Python and the critical functions were implemented in C++ with the parallel loops optimized using OpenMP and wrapped in Python using SWIG. We have a different max-tree structure that is more suitable for array processing, and allows a fast development of new methods with a reasonable processing time. Our main goal in providing this toolbox is to spread the max-tree data structure further among the scientific community, investigate it further, and develop new tools that may be applied to solve real signal, image processing, and pattern recognition problems. This toolbox works for both 2D and 3D images of type uint8 and uint16.
This fork: The code in this fork was ported to Python 3, and is using scikit-image instead of OpenCV. Also, the binaries are compiled with Intel C++ Compiler for OpenMP.
- Roberto Souza - roberto.medeiros.souza@gmail.com
- Letícia Rittner - lrittner@gmail.com
- Rubens Machado - rubens.campos.machado@gmail.com
- Roberto Lotufo - robertoalotufo@gmail.com
This toolbox requires OpenMP, SWIG, OpenCV, NumPy, gvgen, and dot for rendering graphs. In fact, you only need gvgen and dot if you intend to draw graphs.
The documentation of this toolbox is available at: http://adessowiki.fee.unicamp.br/adesso/wiki/iamxt/view/
Download the zip file, unzip it, go to the toolbox folder and type on terminal:
python setup.py install
This is the first time that I try to wrap some code for distribution, therefore you'll proabably need to make a few modifications on the drawing functions to have it working on your machine, specially if you are not using Linux (ubuntu). If you have experience in making distributions and you have tips on how to make them more portable, I'll gladly listen to them!
If you have any doubts, questions or suggestions to improve this toolbox, please contact me at: roberto.medeiros.souza@gmail.com
BSD 2-Clause License. Copyright (c) 2016, Roberto Souza and collaborators. All rights reserved.