I have this CMU cam2+ robotics project that requires image processing, where do i start from??
Author: Matthew Wilson
Question by Divz: I have this CMU cam2+ robotics project that requires image processing, where do i start from??
I think i am pretty much clear with the idea that the cam actually tracks a ball. I would like to know if there is anyone who has already worked on the same and has experience with the image processing part of the project. For example, what color model is better if the cam is used? After the entire robot is set up how do we actually get the data from the robot?
Best answer:
Answer by Snoopy’s Best Friend
You asked a lot of questions here.
In my experience, the best color to use with the CMU cam (speaking from experience with the original CMU cam) is in the red spectrum — red, light red, orange, etc.
The camera has various modes that you can extract data from it. One mode will tell you how much of a color is visible to it at that point in time. You can also extract where the bulk of each color is in relation to the camera frame. That is how you can tell if something is moving, where a certain color object mass is in relation to the camera frame, etc. The camera sees in two dimensions so it can’t tell you how far away something is, but it can give you x-y coordinates in relation to the frame avaialble to it at that time.
If you go to various robotic sites, there are samples of these programs. I am assuming you have already been to the CMU website as well.
I am not quite clear on what you are asking in your last question.
Good luck.
Add your own answer in the comments!
Popular Posts:
- Installing Virtue OLED Board & Laser Eyes in Dye DM9 Paintball Gun
- Bridging Digital and Physical Worlds With SixthSense
- Official Angry Birds 3 Star Walkthrough Theme 3 Levels 1-5
- HTC Schubert
- Hook Your Guitar to Your iPhone and Rock Out with iRig
Tags: 
