Hello. I didn’t know where to post a suggestion for digiKam so that the developers can see it, until I found KDE Discuss! I made this exact same post on Reddit, but I am cross-posting it here so that it gets more caring eyes. Sorry if I am breaking any rules or if I am not in the spirit of this community here somehow, I am just an average user. I did find the KDE Bugtracker where I can apparently request features at too, but I wanted to ask about here before I make a redundant/wrongly formatted bug report.
I am putting this under Brainstorm since it’s a feature I would LOVE to see. But I don’t know how feasible it would be to implement this, even if it’s a simple enough feature on the surface, due to the open-source nature of digiKam.
Also, apologies for my bad English! 
So basically, I just want a small feature where I can mark images, which were detected as duplicates by the Similarity tool, as non-duplicates. I’d like to select this option either for the whole set of images (i.e. the reference image sets shown in the left sidebar with the Similarity tool), or just specific images within each of these reference image sets. It can just simply be a right click > “Mark as not duplicate(s)” option and/or a keyboard shortcut. When you do this, digiKam can basically ignore considering these images for the duplicate detection/similarity tool. It can be similar to how face-tagging is done when can ignore faces, but instead here, digiKam ignores duplicates so that they aren’t found when you press “Find Duplicates” and are not deleted when you press “Remove Duplicates”.
I know some people would just suggest me to increase the Similarity Range minimum percentage. For the type of media I am dealing with, there are vast amounts of duplicates (basically the same images) which are still detected correctly by digiKam even at 75-80% similarity. But since I use 75-80% similarity, there are also several images which are not duplicates and I need to preserve both (or more than 2) versions of the images marked as duplicates of each other.
The current workaround I have to perform right now to make digiKam not delete non-duplicates detected as duplicates is: look at non-duplicate images detected as duplicates (sometimes I need to verify myself if they are actually duplicates by going to the Preview view), select the images by dragging the mouse and highlighting the images in the Thumbnails view, press Ctrl+X for cut, go to my File Manager on a different folder outside digiKam’s control, Ctrl+V to paste images into this folder, go back to digiKam again, press any ref. image set on the left sidebar so I can press Down Arrow key to quickly look through all the other detected duplicates for non-duplicates, rinse and repeat. When I am done, I copy back all of the non-duplicate images back to the appropriate folders. Then I have to go to “Tools” on the menu bar and select “Scan for new items” everytime I do this, or else it messes up some stuff. It’s kind of repetitive and annoying, because it requires so many clicks and mouse movement for every single instance of non-duplicate images I see.
So I beg the developers to implement this feature, and it would make my life so much easier!!!
Please do let me know if you have any questions, or what you think about this!