Essential Software

  1. Python. Use version 2.6 or 2.7, if you are new to python.
  2. numpy
  3. scipy.ndimage

On Windows, use WinPython, a portable Python distribution including a long list of packages from the scientific community. It runs out of the box, and does not need an installation.

If you are using Linux, you should use the packages that come with your distribution. If you are using Mac OS or Windows, check out Enthought's Python Distribution as a good solution if you are not very familiar with how to install Unix-style software. It includes all of the packages above.

  1. ipython
  2. pymorph: Basic (and not so basic) image morphological operators.
  3. scikits.image Includes bindings for the OpenCV library.
  4. mahotas: Several algorithms in fast C++
  5. simpleCV: Computer vision library.

Other Software

Machine Learning

Often computer vision uses machine learning at its core. Some machine learning packages did it all, but it is better software engineering to keep these separate.

  1. scikit.learn
  2. milk
  3. pymvpa

Anything Missing?

If you find any important piece of software missing from this list, you can contribute to this page.