My uncle is a photographer with lots and lots of photoshoots. He stored (duplicates of) backups across a variety of external hard drives. 90% of all backups are the same, but some have had some edits, such as added files or photoshopped images. Now I got him a NAS with RAID setup and an off-site NAS for additional backup of the main machine, to store all his data on one single device. He asked me to deduplicate all backups and only keep the latest version, deleting all copies.
There are already a few applications (sure, recommend one if you know one (Mac/Windows) to identify copy images when file size & file name are equal. But I was wondering..
TL;DR
What are the odds of an image to keep the same file size when it's photoshopped?
//
Alternatively: What's the best way to identify perfect duplicates of images?
In the case my previous work method (filename/size) isn't the best way to go at it.
I had the assumption that if the odds are very low (<1%) I could simply search & destroy and keep 1 of every file of which both file name and file size is equal.