CZI: Incorrect Detector Gain
The issue appears to be that the detector gain metadata is incorrect for the first detector. This was consistently reproducible and a sample file was provided that reproduces the issue and is available at scratch/inbox/imagesc-80767.
In this sample the first detector always has a gain of 1. The correct values should be as below:
The sample showing the issue was acquired with the older ZEN black which was afterwards processed in ZEN blue.
An attempt was made to fix the issue in https://github.com/ome/bioformats/pull/4002 This however resulted in incorrect and modified behaviour for existing datasets, WF.czi, SIM_nicht verrechnet.czi, and SIM_verrechnet.czi
In these cases we are first populating a list of gains from the Voltage node of Detector, then we are later reading the gain node. Finding the true source of data seems to differ between datasets and so finding the right value for all the samples we have has proven difficult.