All geek questions in one place

OpenCV packaging with python application

So, I am considering which language to use in developing an application using OpenCV. As part of my decision, I am interested in knowing how easy / difficult it is to include the opencv library in the final application. I would really like to write this in python, because opencv bindings are great, simple python, etc.

But I couldn’t find a clear answer to such things as “py2app automatically binds opencv when it sees the import string cv” (I think not), and if not, is there a known way to do this

In general, I would like to know what is the best way to distribute a python desktop application with opencv.

+6
source share
2 answers

I have effectively packaged, deployed, and shipped a Python application using OpenCV with cxFreeze.

http://cx-freeze.sourceforge.net/

Yes, cxFreeze automatically raised python extensions on OpenCV. I had to manually copy the OpenCV DLL (Windows), but this was a small problem that could be solved at the post-processing stage in cxFreeze. It picks up other DLLs, so I'm not sure what the problem is.

In fact, it worked so well that I was surprised. The only thing he did not pick up properly is the QT library. Basically, here were my steps (I use QT, so ignore this part if you haven’t done so):

  • cxfreeze App.py --target-dir App --base-name Win32GUI --include-modules PySide.QtNetwork
  • Copy the opencv \ build \ x86 \ vc9 \ bin files to the application directory.

What is it.

+1
source

You can take a look at http://www.pyinstaller.org/ This is great, and it packs the necessary dll for opencv.

For opencv to work, this py2exe is also a numpy package. Make sure you add a dependency for it if you are using py2exe.

0
source

Source: https://habr.com/ru/post/920455/

More articles:


All Articles