I presented a workshop this afternoon at workshops hosted by Team 1418 in Falls Church, VA. The topic was “Image processing on the driver station using python”, and discussed some basic image processing techniques that my team successfully used last year. It was pretty well received, so I’ve posted the slides + sample code on my website. I hope you will find it useful!
Dude, you’ve just helped get rid of some of my headache! Reps to you! 
I have been wanting to do onboard vision tracking and my team wants to do DS tracking. However, it becomes a lot harder with the lag, so we would need much better programmers on our team than we have. However, I would need to do this in C. Anyways, Thanks!
Thanks for uploading this. I definitely want to try it out at some point!
Is it true that opencv needs to be used with python2 rather than python3? (We have it working now for python2).
Our team is also learning a lot from using your fake-wpilib and had been using that with python3. So we are experimenting with having the robot code developed and tested on a laptop with python3, with pynetworktables configured for python3 to aid in testing communications.
Then we’re experimenting with building a driver station vision processing python2 app, using opencv (and also standing on the shoulders of Team341.) So again we configured pynetworktables to work with python2 as well.
Thanks for blazing the trail.
That is correct. I believe the next major version of OpenCV (version 3.0) will have Python 3 support, but it has not been released yet. According to their roadmap, it is expected to be released in February 2014.
Thanks for blazing the trail.
Glad to hear someone else is finding it all useful! 