3
\$\begingroup\$

I am using digiKam, and I have loaded all my photos. I have several duplicates, sitting in different disks/folders.

I discovered the "find duplicates" function, and the "remove duplicates" function, but I don't understand how this last works: which is the criterion used to identify the reference image? (i.e. the copy that is kept, while the others are sent to the wastebin) Is there a way to change this criterion?

I read the manual, and only found a short description of the function, without the details I am asking. Also a google search did not give any more details.

I have digiKam version 7.4.0, in Linux Mint 20.2

Improve this question
\$\endgroup\$
4
  • \$\begingroup\$ Did you try searching the web and/or manual? If so, could you include in your question what things you've searched for, found and were not helpful? That narrows down the options others might suggest. \$\endgroup\$
    – Saaru Lindestøkke
    Jan 10, 2022 at 20:37
  • \$\begingroup\$ I shortened the question, and made it clearer (I hope). \$\endgroup\$
    – Fabio
    Jan 12, 2022 at 8:14
  • \$\begingroup\$ Clarity is not the issue with this question. It's perfectly clear what you want to know. However, it is not clear what research you've done yourself. Did you look in the manual? Did you look online? What did you find? If what you found was not helpful, why not? \$\endgroup\$
    – Saaru Lindestøkke
    Jan 12, 2022 at 10:11
  • \$\begingroup\$ I see your point. I added some description about my research and attempts :) \$\endgroup\$
    – Fabio
    Jan 12, 2022 at 11:24

1 Answer 1

Reset to default
1
\$\begingroup\$

Searching for "digikam find duplicates reference" on the web I found this Reddit post (titled "Digikam Reference Photos - How are they determined?") as the third sub-result of the first hit.

A commenter refers to a related feature request about the selection of the original in duplicate search. This feature request is closed with this commit in the 7.4.0 version (bolding mine):

use the oldest image date or larger pixel/file size as the reference image
So far, the reference image was actually more of a "coincidence", depending on the image ID in the database. That makes little sense, because we are looking for the original image if possible.

From the code it doesn't seem that this behaviour is configurable, but I might be mistaken.

Improve this answer
\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

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