Fiducial camera weirdness

Started by arvydas, July 25, 2016, 07:50:45 PM

Previous topic - Next topic

spiyda

Quote from: Mike on August 09, 2016, 04:59:02 PM
Or maybe just on a sheet of white paper, assuming you can get the lighting right. A temporary ring-light round the camera maybe ( on mine the camera light is pretty ineffective, but could just be aging LEDs.

I think you have it there Mike, you and I both have a trillion LEDs to play with, we should be able to get a nice even illumination. I have an A3+ laser printer and some white card, I'll give it a shot later in the week and see if the camera can consistently find the marks.

It might be just that easy !

ps,   I did some tests with part of a grid laser printed, and the onboard LED ring was definitely insufficient to illuminate the dots enough.  17 year old LEDs need replacing !

Mike

I'm not sure a laser print would be accurate enough, but certainly worth a try to establish the method, and get a photoplot or typesetter print done once the method has  been figured out.

Mike

Quote from: SteveW on August 09, 2016, 08:06:07 PM
Quote from: Mike on August 09, 2016, 04:59:02 PM
PCB is one option, but may be tricky to get uniform lighting across the whole area. Also PCB doesn't have optimal contrast.

But PCB might be consistent with actual PCBs. Something with massive contrast and backlit may deliver crisper pics, but not line up with where the fiducials appear to be in real use / real lighting ?


The issue is that the grid cal is an automated process that takes many hours, so needs to be able to be left to run by itself, and not be skewed by angled lighting causing offsets.
Lighting differences may make a difference to blob size, but shouldn't afect centering

spiyda

Quote from: Mike on August 11, 2016, 02:39:13 PM
I'm not sure a laser print would be accurate enough, but certainly worth a try to establish the method, and get a photoplot or typesetter print done once the method has  been figured out.

At least we know the camera can use a laser printed target properly once the lighting is reasonable.

I have a 2400 dpi A3+ laser which is pretty accurate, very linear and certainly repeatable and I can print and then compensate, so the limiting factor will be my ability to measure, and the accuracy of what I use to measure.

In aluminium, the calibration plate will expand about 0.1mm in 5 C  (steel would be about half that)
and getting it cut by a commercial laser will give me datums accurate to about 0.1mm ( so they tell me) to align the printout to.

Cutting all the registration dots as holes gets expensive and we don't really know how those align with the plate until we try.
So a cheap laser printout on card to play with would at least be a start

At this stage I don't know if that is accurate enough, but I'm inclined to get an aluminium base plate made up to play with.

Has anyone on the forum actually seen the original jig?



spiyda

A mild steel jig is being made (lower coefficient of expansion than aluminium)

I'll report when I have had a play !