I use the Image Reader to read an CZI-File (2 Channels, 20 Time Frames). This seems to work OK using the BioFormats Reader (it works perfect in Fiji or ImageJ), even if the output port only shows black images.
My next step was to use the Spitter. I selected XYZ and Time as dimensions to end up with two images, one for each channel (also black).
And finally I wanted to use Image Calculator to divide CH1 by CH2, but the Console showed:
WARN DetectorTypeEnumHandler Unknown DetectorType value 'null' will be stored as "Other"
WARN CorrectionEnumHandler Unknown Correction value 'null' will be stored as "Other"
WARN MediumEnumHandler Unknown Medium value 'null' will be stored as "Other"
WARN BinningEnumHandler Unknown Binning value '1,1' will be stored as "Other"
WARN BinningEnumHandler Unknown Binning value '1,1' will be stored as "Other"
ERROR . Execute failed: Required service is missing: imagej.data.DatasetService
1. the problem that the output port only shows black images:
the images you read are probably not normalized, i.e. the pixel values do not cover the whole range of the pixel domain (e.g. if its an UnsignedShortType image, pixel values can range from 0 to 655535, but probably the pixels in your image only take values way below that maximum value). So, you can either view the images by opening the Table Cell View (e.g. in the Image Reader context menu or the Table Cell View node) or put the Image Normalizer- or Image Converter (with the Normalize and Scale-Option) after you read the images.
2. Execute failed: Required service is missing: imagej.data.DatasetService:
You were appearently using the Image Calculator...-node of the imagej2 intregration which still is quite experimental (so is ImageJ2 itself). Hence, I would recommend you to use the Image Calculator (category: Community Contributions > KNIME Image Processing > Image > Process, or just search) of the stable Image Processing Extension for now and we'll fix the ImageJ2-Integration-issue. Or doesn't this node meet your requirements?