View Full Version : Axis Cam brightness in LabView
Chath Maral
07-02-2009, 13:17
Okay, we've been messing with the hue, saturation, and lumination of the colors to get the pink/green combo of the flag registered for awhile now.
So far, we realized that the 0-255 code doesn't work, but we can't locate the Iris control on the camera (different from manually adjusting the focus, which we just found) which will be crucial since the field will be mainly vividly contrasting darks and lights.
Greg McKaskle
07-02-2009, 15:17
I'm not sure what you mean about the 0-255 code doesn't work. Can you be more explicit?
As for iris, the camera has a fixed iris. And yep, the focus is manual.
What problem are you having?
Greg McKaskle
Chath Maral
07-02-2009, 15:58
Well, the coding for the 'pink' is correct, but it also picks up magenta and mainly anything red that has light shining on it.
The 'green' is also correct, but it is too specific and only show a few particles.
When I attempt to mess with the green, I get no picture at all.
What I mean by no 0-255 is that the Html color code I am trying to use in LabView doesn't seem to have set parameters. It goes into the negatives and beyond 1,200.
In fact, I belive you were the one who's code I used from http://joule.ni.com/nidu/cds/view/p/lang/en/id/1215
Greg McKaskle
07-02-2009, 19:47
The mask for both of the colors is set pretty wide, so as you noticed, bright orange, red, and magenta will get picked up too. That isn't so bad.
You say the green doesn't look so hot. The best way to determine which parameter is off is to open up the subVI called FindTwoColor. On its panel is a button for debugHSL and an image display which you can mouse over to see individual pixels in HSL. This is off by default because it slows stuff down, but for tuning, turn it on and hover over the pink and green areas. Compare that to the HSL numbers on your top panel.
Ideally the H number is somewhere near the center. It probably needs to be around five or ten from the high or low.
The saturation needs to be above the minimum by ten or so, and the luminance above the minimum by ten or so.
If your light is sufficiently different, you may need to bump the numbers a little. You can do it while the VI runs. The other option is to turn on more lights or move to a place with more even lighting.
I would really like to hear which parameter was off and how much you had to change it to work.
Greg McKaskle
vBulletin® v3.6.4, Copyright ©2000-2017, Jelsoft Enterprises Ltd.