I want to merge two grayscale images into one, ideally taking the average of the two images. Both images are grayscale, same canvas size, flattened. I tried using merge channels but that puts me into a new image with two channels.

I'm using photoshop 2021 if that matters.

  • Can you define what property should be averaged? Should it be grayshade pixel by pixel - for ex. grey there where one has black and the other has white? Or should the content be recognized in some higher level classifying system? For ex one image has a horse and the other has a donkey, the result should be a mule? – user287001 Apr 3 at 6:28
  • Yiu have to be a bit careful when you start to do mathematical operations on a image. A lot depends now what the image is like. The 50% layer does average, but its only correct if you assume the data in your image is linear. Generally thats not true. – joojaa Apr 3 at 6:35
  • Hi. Welcome to GDSE. Can't you use a layer blending mode, such as overlay, multiply, screen, etc? – Billy Kerr Apr 3 at 9:56
  • @user287001 yes, grayshade pixel by pixel. – Astro.nor Apr 4 at 20:16
  • @joojaa the image is of Messier 64, the evil eye galaxy. The two images were taken with filters, so the wavelengths of light are roughly 30 nm apart. This image would take the place of the green channel in a later image, but it was an extremely short exposure compared to the red and blue images, so I want to combine 2 images that are green(ish) so I have a more balanced final image. – Astro.nor Apr 4 at 20:19

Set the top layer as opacity 50% in a document with each image in it's own layer.


This is only a too long comment:

You have got a perfect answer from user Alexander Samual Dunnett for the presented question. Unfortunately the question didn't present what you want. User joojaa guessed it and wrote a warning.

The comments revealed that you want to combine 2 grayscale photos to get one which is a substitute for a photo which was taken with longer, say double length exposure.

We do not have a slightest idea what camera system you had when you took the photos and what mangling the photos have got between the exposure ending and being "ready to be combined in Photoshop". Only summing the original unmangled data can be considered as something you expect. Revealing what there originally was collected is impossible if images are processed to good looking 8 bit/channel usual photos - a vast amount of original info is thrown overboard without asking because it was probably considered to be only noise - too weak to be interesting - and non-linear dynamics compression has mixed the rest together in a non-reversible way.

The only way to extract something which is now invisible is to start from original RAW images which have got no processing. That's the data what you have.

To get what you expect you need other software than Photoshop as said by joojaa and Billy Kerr - the latter even picked one freebie (=DeepSkyStacker) which at least knows how to use RAWs. Unfortunately I do not know does it solve your current problem.

For more help contact people who have worked with the same problem. They very likely can be found among astronomy and astrophotography hobbyists. Of course also some of us here in GDSE are able to present things which are normally invisible, but they make art.

  • There's also Photography Stack Exchange, and there's an astrophography tag. I see quite a few questions there already on stacking. Although I suspect the OP would need to add much more information, possibly including sample photographs, in order to get some help there. – Billy Kerr Apr 5 at 10:23

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.