Gray scale calibration, always get an error....

[SolrunSig]

Hi,

I am using Zooimage at my institute and I have never had any problem with grayscale calibration.
I was scanning the filters the other day and when I wanted to calib() then I always got this messages....

calib()
Calibrating gray scale... [cal_2.tif]
WhitePoint=0
BlackPoint=51868

Take care:
Images are overexposed, or whitepoint is already calibrated
Wrong OD calibration: not a straight line relation at alpha level = 0.01
Bad OD calibration (R squared = 0.976)

How can I fix this, through the years I sometimes have had to scan the filters more than once to get the white point =0 and only get the message of overexposed Images but not the OD calibration.

It would be really nice to get a answer soon as possible :)

[SolrunSig]

I am using version 5.5.2 of zooimage.

[phgrosjean]

The first message "Images are overexposed, or whitepoint is already calibrated" is not a problem when whitepoint = 0.
The two other messages indicate that there is not a linear progression in the ODs. Are the different gray areas relatively homogenous in the picture ? Are there excessive shadows in the image ?

If one calibration image is not usable, but the other ones work well, substitute the problematic image with another calibration image that has similar gray values (you can check this by opening both images in ImageJ and looking at the values for the gray areas, or by comparing the respective grayscale histograms).

[SolrunSig]

I already looked at some images that I have used for calibration for previous surveys, and what I see is that the values seem to be a bit higher now than previous images that I have used for other surveys. I always do a new grayscale calibration when I begin to scan samples from a new survey. The the first two histograms are from images where the calib() went fine (since end of 2021)but the third and fourth shows histograms from images scanned this week (already scanned up to 40 images and I always get the same message).

[phgrosjean]

The middle gray filter seems to be more heterogenous. This is already the case in the two first images, but it seems a little bit worse on the two last ones (red marks). You should check and clean, or replace that gray filter may be ?
It should be possible to change thresholds that trigger error messages during calibration so that you still got (approximate) values for whitepoint and blackpoint. But it is certainly better to solve the problem instead.