Removing color banding from a 50-year-old polaroid

[drkddell]

Hi everyone-I'm an experienced photographer but relative novice to Photoshop, at least with respect to picture restoration. Attached is a downsized image from my scan of a photograph provided by my parents for restoration/archiving/reprinting. As you can see, there are vertical orange stripes in a regular pattern across the image, which are definitely present in the original photo, not scanning artifacts. Any ideas how to deal with them?

[Chuck Uebele]

You can try using color range to select the very red/yellow areas to make a selection, then use that selection with curves to reduce the red. You will have to do this a few times to get all the offending areas, but here's a start of what you can do:

If parts go too cyan/green, you can paint those out on the adjustment layer mask.

[Lucas Lago]

Hello drkddell. You need to select this orange areas, and start to working to match colors. But each area will need your own selection and changes. Like one layer to correct the Left Trees, another to correct street, another for legs, grass, etc.

I would start using the eyedrooper tool and picking a color next around the area you want restore, full fill the selection in a the layer and try using overlay modes like Hue/color/saturation. If the result is not so good make a mask of the area on original image and try use use some colors change in menu>images> color correction /change colors/ etc.

You can use both methods together too. There are several ways to do this, but I think this is a good start! Remember that a good selection and feather is much important as well.

[Stephen Marsh]

Would the following very quick example be a suitable start point? The idea being to make a very fast correction that will hopefully create a suitable base before more time and effort is put into the image.

If so I’ll detail the steps or put it into an action so that you can get to a similar point and then put in more time and effort. For me, the question always starts with – what do the individual channels look like, and how can a “good” channel be used to help restore a “bad” channel?