How to set threshold for image J and then do image calculations?

Hi everyone.
I am now using Image J to do image calculations, using one image series to divide another image series. However, the two data might have only small difference or the values of some parts (voids) of the two data are closing to zero. therefore, the divided results showing only value=1.0, no details are shown. Someone else suggest me make a threshold for the data first to remove the void parts. Could you please tell me how to set a threshold to the pictures and then do the calculations? I want to use image w to divide image e. but i got nothing before setting threshold to them. Thank you very much!

Dear @amy_ustb,

you can generate a binary image, i.e. an image that has only pixels with values 0 or 255, if you click the Apply button in your screenshot.

You could then divide the resulting binary image by 255 to get an image of 0s and 1s. Multiplicating this image with the input will generate an image where below threshold values are 0 and others keep their original value. For that purpose, however, it be easier to use the Image Expression Parser that is for example bundled with Fiji.


Hi @stelfrich,
Thank you very much for your suggestion.
I followed your advice, however, when I scroll the slide on the threshold panel, there are no changes for the values in the circle. That means that the upper and lower value of the threshold is the same as 0.
could you please give me some tips on this?
And I am now using Fiji to try this.
Thank you very much!

All the best.

Which circles and which values are you referring to here?

Click on the image to see the full screenshot.
(I tried to edit the post, but there is a display issue with the forum software apparently, which leads to the image always being displayed at 100% and the specified dimensions being ignored.)

Thanks for pointing that out, @imagejan! :thumbsup:

If nothing went wrong, that’s what those two values are denoting. How exactly did you produce the image that we can see in the screenshot, is it still the original image?

Hi@ stelrich,
Thank you very much for your help.
After I import the original Image picture, I just used Image–threshold, and then the red colored parts occurred. But, as you have seen, the value of the threshold still remained as 0.
But I think @Johannes_G might have found the solution to this problem.

Many thanks.