# How can I convert a stack of images with increasing intensity to a scale that will let humans see the difference in intensity?

I have a plethora of tiff stacks from a camera with a CCD attached, I am observing the increase in the intensity of my image stack over a change in concentration, the problem that I am having is that this change, although clearly shown on my graph of the average pixel value of these images, but when given the most and least intense images side by side this progression is nearly impossible to see.

I am wondering if there are any ImageJ functions, or other free programs, that will convert these intensity changes so that the difference is something that can be seen by the human eye.

• Could we see a couple of these that illustrate your question? – Stan Jul 10 '18 at 22:23
• Does he talk about HDR stacking/bracketing? – Horitsu Jul 11 '18 at 5:12
• In a previous question, he mentions "using a fluorescent microscope with a CCD". I'm guessing this is the same set up. – xiota Jul 11 '18 at 12:50
• When you say stack, do you mean a an MSI / HSI hyperstack or a 3D (x/y/_/t) stack? Are you trying to measure total contribution (I.E. cummulative histogram) or trying to measure Delta H (I.E. change over time?) ImageJ can definitely do what you need but we require more detail about your process. – PhotoScientist Jul 13 '18 at 13:26
• @PhotoScientist I'd be interested in reading about both of those if you write it up... – xiota Jul 13 '18 at 23:55

I don't understand what you are doing or describing, so this answer will likely be revised if more information becomes available.

Within a single image, if there is a gradient that is difficult to see, we can expand its range using levels or curves. For example, if we have a 10-step gradient from 45-55, it may be difficult to see the difference between 50 and 51. But if we stretch it out to span 0-100, it's easier to see the difference between 50 and 60.

It seems you want something that will do this with a "gradient" of images instead of pixels?

• You may be able to write a script that calls ImageMagick (perhaps changing brightness, contrast, or gamma). Whether this will work depends on the nature of the differences you want to make visible.

• If the images will be displayed on a monitor, you can use an interface in which the user can compare images by switching back and forth between them.

• To graphically illustrate the differences without manipulating the images themselves, you can show a "deck" of thumbnails spread along a gradient with callout to a larger image with key areas circled and captioned. Then be sure to clearly state, "For illustration purposes only."

• Based on your previous question about CCDs and gamma, if the differences are difficult to see because they are represented linearly, it may be acceptable to gamma correct them, as long as you note that this is what you've done. (For instance, if you are photographing electrophoresis gels.)

• It's worth noting that if the images are intended for scientific publication, image manipulation may not be acceptable. If the only difference is the relative "brightness" of the images, manipulating the images could be seen as dishonest because the manipulation alone would have resulted in the brightness difference.