Really not sure what the problem is, but I’m so close to getting it work.
The connectivity problems have been solved, thanks to suggestions from other users here, but now I’m having difficulty getting a clear image of the slide material.
The image comes up as a bright green, and I’m not sure what the cause of it is. I’ve tried disconnecting and reconnecting the Pi2 camera, on both ends, and that didn’t seem to make any impact.
Even after unscrewing the 40x microscope objective, the image is still green.
But the camera seems to be detecting the material on the slide, displaying shadowy globs here and there within the green.
Is there an optimal material to put on the slide for testing? Right now I’m using saliva, but maybe that isn’t the best choice.
Also, the piece that the pi camera is mounted to was printed with white PLA, which seems to let in some light. Might that factor in to the problem? For the time being, I’ve wrapped it in black electrical tape, because I was also experiencing difficulty in printing out another piece in black PLA. The base of the STL seems to be easily ripped apart by how quickly the nozzle on my printer moves, so I might just spray-paint the white piece black, if needed.
This issue ,I have also faced and keep facing.I keep autocalibrating using camera again and again and sometime by chance gets corrected on its own.The probability increases whenever I calibrate more than once…though I don’t know the logic.
Yeah I’m able to zoom in an out, but the screen is still almost completely green, seemingly no matter what sequence of focus/adjust/calibrate I try.
When you say it occasionally corrects itself, do you mean the color of the image is corrected? Or is color not your problem?
I am not sure about the green. Having the optics holder in white will let in stray light. Do it black if you can.
I think a better sample slide will help to separate any different issues. Saliva is not high contrast and it is hard to know whether you can see anything. Prepared stained slides are really good and pretty easy to get hold of. Unstained slices of plant are pretty transparent and again it can be hard to see. Not great if you are not already confident in your microscope. A few hairs gives high contrast black lines to get an idea of the focus.
1 Like
I have a H&E stained section of brain. It should be generally be pinkish.However I have been struggling to calibrate the colors.Its been almost two days,I have been repeatedly trying to get it back to normal, however I have failed.I have attached the screenshot of the image.
That image shows that the setup and focusing is giving a sharp image. The green cast is very strong. I am clutching at straws here, but I would try removing any colour calibration, or reinstalling the software from scratch. Somewhere a colour correction has gone mad.
I assume you can get a sensible colour picture with the Norma Pi camera capture software?
Thank you.I ll give it a try.
Installed the raspian openflexure afresh…got back my image…working fine…Thanks @WilliamW
1 Like
I’ll give that a shot as well.
The green colour cast during calibration seems to pop up for a lot of people, and I’m currently trying to pin down why; unfortunately, all my test setups don’t show it! I’d be very interested in seeing the output from running ofm log on your microscope if you have recently done a calibration that went green.
Hi, did you check the RAW data of the image? Maybe the camera sensor isn’t working properly.
I realise I mentioned this in another thread, but it’s worth linking to it here: the current beta release (2.10.0b1) includes a rewritten calibration routine. This seems to fix the white balance issue 
1 Like
I am having issue with autocalibration as well, it goes green and will never be back to normal. Have to reinstall the software again.
From the screenshot it looks as though you are using the server version 2.9. The green image is a known issue with that server version. It is particularly likely to happen when there is low light at the sample, and as you say it gets stuck green and does not resolve easily. You may not need to completely re-install, using
sudo ofm reset
sudo ofm restart
will clear most of the calibration from the server and can give the calibration a fresh start.
The issue is resolved by a new calibration routine in the next version of the microscope server software. That is currently a beta release. Release notes and how to install are on the threads Software Beta release (the Beta1 that has been out for a while) and Software release v2.10.0b2 a beta2 release with some extra functions.
sudo ofm update
sudo ofm upgrade --pre
will give the current latest pre-release (beta2). If you want beta1 which has been used more there is a way to do that using the actual version number, but I am not sure of the syntax.
Edit: layout of server commands
Hi WilliamW, Thanks a lot an i will try what you just mentioned.
@jenovauh did you get anywhere with this?
Sadly no, not sure why the auto calibration will end up with green image. Everything i see is green color lol. I tried 3 camera and i really dont think it is the camera issue. I also have reinstall the OS a few time and always end up the same.
Could it be my white LED too bright to cause that to happen? it is just a normal 5mm straw hat LED which i doubt so.
Did you try upgrading with
sudo ofm update
sudo ofm upgrade --pre
On the server version 2.9 version it is too little light that usually makes it go green, rather than too much light.
Yes I did what you mentioned.
Still go green for me, maybe i should change a brighter 1W LED and see what happen.
