How to Find and Delete Duplicate Photos on Any Device

How to Find and Delete Duplicate Photos on Any Device

Duplicate photos are the single biggest source of wasted storage in most people's libraries. In a typical collection that's been copied from a camera to a laptop to a drive to the cloud over fifteen years, somewhere between 30 and 50 percent of the files are copies of something already there.

Clearing them is also the least emotionally taxing cleanup you'll ever do, because you're not deciding whether a memory is worth keeping. You're just deleting the fourth copy of it.

Here's how to do it without accidentally nuking originals.

First, understand the two kinds of duplicate

This distinction matters, because different tools catch different things.

Exact duplicates are byte-for-byte identical files. Same image, same size, possibly different filenames. IMG_0432.jpg and IMG_0432 (1).jpg and beach_trip.jpg might all be the same file. These are completely safe to remove — there is genuinely no information loss.

Near-duplicates are visually similar but technically different files. A full-resolution original and the compressed version that came back through a group text. A photo and its lightly edited version. Three burst-mode shots taken half a second apart. Removing these requires judgment, because the versions aren't equivalent.

Start with exact duplicates. That's where the storage savings are, and there's no risk.

The golden rule before you start anything

Back up first. Not because duplicate finders are dangerous, but because you'll be moving through thousands of files quickly and one mis-click on a "select all" can do real damage.

Copy your photo library to an external drive before running any deduplication. If it already lives in one place only, that's a bigger problem than duplicates anyway.

Second rule: move duplicates to a holding folder rather than deleting them. Give it a name like DUPES-DELETE-AFTER-2026-09 so you know when it's safe to purge. Wait a month, confirm nothing's missing, then empty it.

On iPhone and iPad

Apple built this in, and it's genuinely good.

Open Photos → Utilities → Duplicates. iOS scans your library and groups exact and near-exact matches. Tap Merge on a group and it keeps the highest-quality version while combining the metadata from all copies — so if one copy had location data and another had a caption, you keep both.

A few notes:

  • The Duplicates album takes time to populate on a large library. If it's empty, leave the phone plugged in and locked overnight and check again.
  • Select All → Merge at the top of the screen handles everything at once. This is safe, because merge keeps the best copy by design.
  • Merged photos go to Recently Deleted for 30 days, so there's a built-in undo.
  • This only covers what's in your Photos library. Duplicates sitting in the Files app or in third-party cloud apps are invisible to it.

On Android

There's no universal built-in tool, but Google Photos covers most of it.

If your photos sync to Google Photos, it prevents most exact duplicates automatically — uploading the same file twice doesn't create two entries. What it won't catch is the same photo that's been re-saved or re-compressed, since that's technically a different file.

For those, Google Files (the file manager app) has a Clean tab that flags duplicate files and junk. It's basic but it works, and it's free.

For a thorough job on a big library, it's usually easier to copy everything to a computer and run a proper desktop tool.

On Windows

Windows has no built-in duplicate photo finder. You'll want a third-party tool, and there are solid free options.

Look for one that offers content-based comparison — sometimes called hash comparison or checksum matching — rather than only filename matching. Filename matching will miss most of your duplicates.

The workflow is the same regardless of tool:

  1. Point it at your photo folders. Include every location: Pictures, OneDrive, Desktop, Downloads, any external drives.
  2. Set it to compare by file content, not name or size alone.
  3. Run the scan. On 50,000 files this takes 10 to 45 minutes depending on drive speed.
  4. Review the selection rules before confirming anything. Good tools let you set a preference like "always keep the file in the oldest folder" or "always keep the largest file."
  5. Choose move to folder, not delete.

On the keep-rule question: prefer the largest file size when versions differ. Larger almost always means less compression, meaning it's closer to the original. Prefer the oldest date when sizes are identical, since it's more likely to be the original import with intact metadata.

On Mac

Photos on macOS has the same Duplicates feature as iOS, in the sidebar under Utilities. Same behavior, same merge logic. Start there.

For photos outside the Photos library — loose files in Documents, on external drives, in Downloads — you'll need a separate tool. The same criteria apply: content-based comparison, move rather than delete.

One Mac-specific trap: if you use iCloud Photos with Optimize Mac Storage turned on, many of your photos exist locally only as small previews. Running a file-level duplicate scan on the Photos Library package in that state can produce confusing results and, in the worst case, damage the library. Don't point third-party file tools at your Photos Library package. Use the built-in Duplicates feature for anything inside Photos.

In Google Photos

Google Photos won't let you upload the exact same file twice, so your exact-duplicate problem there is usually already solved.

What it doesn't solve: the same photo uploaded once from your phone at full resolution and once from a computer after being resized. Those are different files, so both stick around.

There's no built-in duplicate tool. Your options are to sort visually by date and clear obvious repeats manually, or to export via Google Takeout, deduplicate on a computer, and re-upload a clean library. The second is a real project but it's the only thorough option.

Handling near-duplicates and burst shots

Once exact duplicates are gone, you'll notice the other kind — five nearly identical shots of the same moment.

The efficient approach is not to evaluate them individually. Instead:

Sort by date and scan in grid view at a large thumbnail size. Bursts appear as visually obvious runs of near-identical images. Click the first, shift-click the last, deselect the one or two you want to keep, delete the rest. This goes surprisingly fast — often several hundred photos in twenty minutes.

Be more forgiving with people than with things. Five shots of a sunset: keep one. Five shots of a group at dinner: keep two or three, because in at least one of them somebody's blinking and you won't know which until later.

Leave text-message versions alone if you're unsure. A photo received by text is lower quality than the original, but if you never got the original, it's all you have. Check whether a better version exists before removing the compressed one.

What to check when you're done

Before emptying that holding folder:

  • Compare your total photo count before and after. A 30 to 40 percent reduction is normal on a first pass. A 70 percent reduction means something went wrong — check the holding folder before deleting.
  • Spot-check five or six specific memories you care about. Search for them. Confirm they're still there and still openable.
  • Verify your backup ran after the cleanup, so your backup reflects the cleaned library rather than only the old messy one.

Then wait the month, and empty it.

Keeping duplicates from coming back

Duplicates accumulate for predictable reasons, and each has a fix:

Multiple import paths. If your phone uploads to both iCloud and Google Photos, and you also manually copy the camera roll to a drive, you'll generate copies forever. Pick one primary path.

"Save to Photos" from messaging apps. Every time you save a photo someone texted you that you already have, you make a copy. Save selectively.

Copying instead of moving. When you pull files off a camera card or old drive, you're right to copy rather than move — but then delete the source once you've confirmed the copy is good and backed up. Otherwise the source becomes a duplicate breeding ground for the next decade.

Editing in place without a convention. Every "save as" during editing makes a new file. Decide whether you keep originals plus edits or edits only, and apply it consistently.

Handle those four, and a yearly dedup run will take fifteen minutes instead of an afternoon.


Next in this series: Photo File Formats Explained: JPEG, PNG, HEIC, and RAW