Robotics : 21295


Robotics / 21295	21294 \| 21296

Subject:	Vision Command and Linux
Newsgroups:	lugnet.robotics
Date:	Thu, 4 Sep 2003 20:48:10 GMT
Reply-To:	nanobapt@wanadoo.NOSPAMfr
Viewed:	1383 times

Hi, I've test some Video detection with the Lego VC. And here is my report : My program use the videodev API and make is video detection in "real time" My raw image is here -> http://nanobapt.homelinux.com/img/VC/plateau.jpg I apply a sum of filter (and join color value to the program) and I found this picture : http://nanobapt.homelinux.com/img/VC/detect.jpg I apply a mask too ... to say that I don't want to detect robot in this area : http://nanobapt.homelinux.com/img/VC/calib_.jpg And then I detect where the robot is and have this : http://nanobapt.homelinux.com/img/VC/pointparpoint.jpg First point are really poor cause of the camera calibration ! I'll try other method ... Just wait my other result :) note that it will not really hard to make a soft like lego make. Just have some time to write it ! (the same system of patern) have fun nanobapt

Message has 1 Reply:

		Re: Vision Command and Linux
(...) Do you think it would be helpful to have better lighting for the original (raw) image? That might create a better difference between the robot and the background. (22 years ago, 5-Sep-03, to lugnet.robotics)

3 Messages in This Thread:

Entire Thread on One Page:: Nested: All | Brief | Compact | Dots
Linear: All | Brief | Compact

Custom Search