Hello all, this year my team has started trying to work with the kinect and vision processing. We’re doing quite well however we’re stuck on how to identify if a target is hot or not. We’re using the retro reflective targets as means to tell what’s going on during autonomous, however we’re unable to figure out how to isolate the horizontal bar from the vertical one using simplecv and openkinect. Is there anyone who would know of a way to do such a thing?
Once you’ve identified both particles for the vertical and horizontal reflective tape, look at their aspect ratios. The hot goal tape will be much longer than it is tall.
There’s potentially a much easier way to use the Kinect to (indirectly) sense the Hot goal. Hint, read the official Q&A.
You can control the robot with the Kinect during autonomous :P. Is that what you are talking about? :yikes: :ahh: ::rtm::
Bingo. But watch out for the case of seeing both sides of the wall, then you’ll have 3 targets to sort through.
Yes. And lower risk of damaging your Kinect and its easier to power. (A/C outlets at driver station this year.)
Wow, I can’t believe I didn’t think about that. Thank you very much.