Making the thermal scale range smaller in ImageJ

I am trying to represent the fluorescence intensity difference using a thermal scale. However, due to the relatively small difference in fluorescence intensity, I cannot utilise the full LUT range to show the difference in intensity and therefore, the difference appear the be quite small. While this means that there IS a small difference in intensity, I would like to know if it is possible to modify the LUT range so that the thermal scale covers a smaller intensity range. In other words, is it possible to delete the high and low ends in the thermal scale LUT and readjust the thermal colours to span within a smaller LUT range? Hope this makes sense…

1 Like

Hi @koku003,

have you tried the menu Image > Adjust > Brightness & Contrast?

Hi there,

Thank you for your reply. Yes I have. However, this does not change the actual range of the thermal plot. I can modify the colours quite well though!

What I actually want to do is to modify a particular LUT which is in somewhere in the middle of the scale to be the min/max LUT and have a thermal scale starting from there. Not sure if this makes sense!


For example, this is a thermal plot and there are some yellow parts there, I would like to modify the thermal range so that those yellow spots are red, without changing the brightness, contrast.

When I drag through the region I want to change in Edit LUT and change the colour, it does not seem to follow the thermal range of blue - light blue- white - green - red too. the only way I can think is to actually go through them manually one by one!

Sorry, I’m not so familiar with manipulating LUTs. But I’m wondering why you want to do this. Isn’t it illegal from a scientific point of view to manipulate LUTs? How do you justify your manual corrections? Can you think of a better way of normalising visualisation of your data?

1 Like

I agree completely. I have quantified the images seperately to show a graph, however, we are interested to show the difference in a more easy to see manner especially for our timelapse. Perhaps normalising the data may be a way forward.

I think I found a way. I can use Red/Green scale, which has very defined colour difference