How can I identify DNG files that are duplicates of my recovered original RAW files?

Asked 12/8/2020

4 views

2 answers

0

I recovered an old hard drive and found the original CR2 and NEF files that I had previously converted to DNG and kept instead of the camera originals. Now that storage space is less of a concern, I’d like to restore the original RAW files and remove the duplicate DNG versions where they represent the same image.

Is there a practical way to match DNG conversions to their original RAW files so I can identify duplicates safely?

Originally by Photography Stack Exchange contributor. Source · Licensed CC BY-SA 4.0

Photography Stack Exchange contributor

5y ago

2 Answers

0

AI Answer

Generated from our catalog & community — verify before relying on it.

Yes. A normal file-duplicate finder won’t help much because the DNG and the original CR2/NEF files are different files even when they come from the same capture. You need an image/DAM tool that can match files based on image content and metadata rather than exact binary identity.

One approach mentioned was IMatch, which is designed for image matching and digital asset management. Tools like that can compare images by what they contain, helping you find DNG files that correspond to the recovered originals.

Before deleting anything, verify matches carefully. Converting to DNG can preserve much of the RAW data, but the files are not identical, and sidecar edits/metadata may need attention. A safe workflow is:

  1. Import both the recovered originals and the DNGs into a DAM tool.
  2. Use content/metadata matching to locate likely pairs.
  3. Confirm matches manually.
  4. Back up everything before removing any DNGs.

So the key is to use a photo management application with visual/metadata matching, not a simple duplicate-file utility.

UniqueBot

AI

5y ago

0

I use a program called IMatch (from photools.com) to manage my image files (and a few other file types as well). It is called "IMatch" specifically because it was originally based on the author's graduate work on matching images based upon what they looked like, rather than just the arrangement of bits. It does an incredible job of matching images. You can even register some sample images of people's faces and it will identify the people in your pictures (and auto-tag them if you want). It even has a feature where you can literally scribble something that looks vaguely like a picture you remember and it will find the closest matches. It's actually pretty darn good.

It is also a very powerful Digital Asset Management (DAM) tool. It now incorporates EXIFtool for all of its manipulation of metadata, so you know it is modifying the metadata in an industry standard manner. It has no image manipulation features, like Lightroom does (other than basic file resizing). But I prefer to keep those two major functions separate. Instead, it has categorization tools that I feel are much better than Lightroom's. (Of course, I haven't looked at Lightroom in years.)

As I mentioned, it will also manage other types of files. If that file type does not use EXIF metadata, then the program will create "sidecar" files to store that data. Of course, all that data is also stored in the program's database. But the author feels very strongly about always giving you full control of your data, so it mirrors that data to the metadata in the files, or those sidecar files.

The best feature of this software is that there is no frikkin' subscription. It is a one-time cost of about $150, and that is well worth it.

Originally by user82503. Source · Licensed CC BY-SA 4.0

user82503

3y ago

Your Answer