Delta Stage calibration

Hi there

I am having come difficulty calibrating the Delta stage camera. Is is best to calibrate with white light and without the filter cube first?
Any and all advice and help gratefully appreciated.

The camera calibration will try to make a flat grey (uncoloured) background, so that will not work if you have coloured filters for fluorescence.

1 Like

So what intensity of white light is it best to calibrate the DeltaStage with?
Again, many thanks in advance.

@samuelmcdermott has much more experience in using the reflection and fluorescence optics than I do.
In transmission you just use the illumination that you have for normal viewing - the calibration then accounts for the illumination uniformity and the camera. For white light reflection it is basically the same, except that you need some kind of uniform reflective white sample with no structure. That is not so easy to arrange. For fluorescence the colour in the image has less meaning so the calibration is less important.

Are you using it for brightfield or fluorescence reflection imaging? Firstly please make sure you are on the latest server version.

If it’s brightfield, then yes a piece of paper where the sample is will help you get a even camera response using the auto calibration.

If you are using fluorescence, you can either use a fluorescent
slide eg from thorlabs to calibrate thr camera, but I would recommend not doing the white balance calibration, only the flat field and gain/exposure calibration. Alternatively, you can get pretty close by running the flat field correction with the transmission illumination, with the filter cube in, but don’t run the white balance and gain/exposure calibration.

You will then need to set the exposure and gain manually to see your samples. Bare in mind the field of view of the fluorescent images are unlikely to fill the whole sensor even when perfectly calibrated.

Keep us updated on how you get on and feel free to keep asking questions!


Just to be clear, the white paper in place of the sample is useful in reflection mode - in transmission mode it’s probably best to just run it without a sample.

I have not calibrated one in fluorescence for a while, but I think we settled on running the full auto-calibration with white light, and then re-running just the gain/exposure. The flat-field correction will definitely be better for fluorescence if you do it in fluorescence mode with a ThorLabs slide as @samuelmcdermott suggests - though in that case I suspect the unused colour channels will contain very noisy data, and if you use that flat-field correction with white light it will look quite exciting!

Currently there’s no nice way to switch between previously-used flat field corrections, except possibly through the API…