I have got two questions regarding the MaxEntropy thresholding method within the Global Thresholder node.

First: Can we output the threshold itself? I want to calculate the threshold for one image and use the same threshold for a multitude of images.

Second: Can we derive the threshold based on multiple images? That would correlate to the “use stack histogram” option available in ImageJ. I want to calculate the threshold based on all images of one well of a micro plate.


Hi Marc,
the methods in the Global Thresholder node are calculated for each image to find the optimal the optimal threshold for just that image. If you would calculate it on one image and then apply on another one, it would not necessarily work as well. We currently do not support exporting the calculated threshold value. Is there a specific reason you want to use the same threshold on all images?

If you want to do that you first have to merge the images into one using the Merger node and perform the thresholding on that new n+1 dimensional image, take a look at this example workflow on how to do it: Stack thresholding example.knwf (14.0 KB)

Hi Gabriel,

I have to translate an ImageJ macro into KNIME. This is for one of our scientific collaborators and he just wants to do it in that way. I cannot give you an explanation, because I have none. Though, ImageJ offers this option and their might be applications for this.

I will take a look into your workflow, thank you very much!


If you need to do this translation 1-1 you might need to fall back to the ImageJ1 Macro node for this operation, but I think it is really not a good idea to blindly apply a threshold in this case.

I will try to talk him out of this approach! :slight_smile:


1 Like