Copy link to clipboard
Copied
The attached image is from a film camera that had a slightly sticky shutter. Which caused each slide to have an over exposed portion, along with a strip of correct exposure. I have several hundred slides to fix, so I am looking for a simple way to blend both sides to a correct exposure. I have played around with various layers and blending options, but I lack the knowledge to get there.
Thank you for any assistance.
Copy link to clipboard
Copied
Some sort of Adjustment Layer with the proper exposure masked out. I would start with just a rectangular mask, and try to get the adjusted part to match. Once they match. work on getting the mask right using Gradients.
Copy link to clipboard
Copied
@Semaphoric has already outlined an excellent solution. It's far from perfect, but I've mocked up an example of how this could be done.
Copy link to clipboard
Copied
Thank you for the mockup, its very helpful. How about a mask created from the image? Could one be overlayed or inverted somehow to more accurately align the two halves? How could that be done? Otherwise I think it could take hours to align/adjust as it seems to be an imperfect gradient between the two sides. Thanks for the reply!
Copy link to clipboard
Copied
Rather than give up on a project that would take ages to complete and provide no assurance of success, cosider saving the images in grayscale or as duotones. Thay would be far easier (and faster) to modify, would avoid the distraction of damaged color and they speak of a time gone by.
Copy link to clipboard
Copied
That is certainly a good approach, but I would prefer to keep the color as the digital file will be the archive of record going forward. To take your idea another way, how about masks or correcting using LAB channels? The L channel contains the grayscale detail. Could the A-B color channels be used to even the exposure?
Copy link to clipboard
Copied
Yes, I did try Lab but in preparation I looked at CMYK to see whether there was some luminosity mask potential hiding in the channels. The Magenta had possibilities and after modifying it, it was applied in Lab where I could work with tonal scale discrete from color. Although the color channels showed improvement, the whole business too much time with no assurance that other scans might not have a CMYK channel with luminosity mask potential. Not a practical solution. I moved on to duotone.