Hello There , i have this camera ( https://www.amazon.com/gp/product/B00KA7WSSU )
and i have bed ( 1600mm × 1000mm ) , i don’t get perfect result specially in edges of the bed and i don’t know ideal mounting distance
i really need help
…
do i have choose standard lens or fisheye lens ??
That camera is only 1920 x 1080. Assuming that it gives you the full resolution when capturing, the best possible accuracy you can hope for is about 1.5 to 2.0mm - you have about 1 camera pixel for each mm of your bed size, ideally you need two or more for 1mm precision or lower.
If the image looks stretched in the middle or curved, it’s a fisheye lens.
in amazon is written (wide angle)
and now what … i need solve … what’s i do in my situation
is lightburn Cameras are fisheye ??
I recently updated to v0.9 and re-ran the calibration. I didn’t know which selection to make and used the TLAR method to determine that my 90° lens LightBurn camera is standard. Selecting fisheye makes for a very interesting capture/camera view during the calibration process.
The calibration went quite well, with most numbers in the low twenties and a few in the mid-thirties (the diagonals).
There’s very little penalty in selecting the wrong option. If it doesn’t Look About Right, just run it again with the other selection.
I have mine nearly set up with the camera. I had had some discussions earlier about it, 1 megapixel and 170 degree field of view. I built a formed plexiglass window/camera mount to position the camera about 2 1/2 inches above the cover, so altogether the distance from lens to work is about 6 inches.
The fisheye choice on camera type made a difference. Captured images were often interesting and a bit psychedelic.
scores ranged below one with good lighting.
Unknown is what does NAN mean on the score? Says “good” nan, select next.
When finished, In the corner farthest from home, it is off by about 2mm, other areas within one. Pixelated images when zoomed in very close but ok.
What does nan mean?
NAN is “not a number” - usually infinity, and it’s what happens when a computer tries to do something impossible like take the square root of a negative number, divide by zero, or something like that. If you see NAN, it’s not a good capture, so you’ll need to try it again.
I kind of thought that might be the case, but was thrown off by the message of a great score, select next. I did find that good lighting makes a huge difference in the readings, Taking a reading, then adding or deleting extra lighting changes whether it identifies an image or not. As it is, knowing I’m using a substandard camera (for now) I am still getting acceptable results. To be determined is the quality of an image capture.
Thanks again. The adventure continues.
Fred
