How to Find Similar Photos on Mac and Keep the Best Shot

Finding and removing exact duplicate photos is straightforward, but the real clutter on most Macs comes from similar photos: burst sequences where you fired off twelve frames to capture one smile, edited copies sitting next to originals, and re-exported JPEGs alongside their RAW counterparts. This guide explains how to find similar photos on Mac — not just identical ones — and walks you through practical strategies to keep only the sharpest, best-exposed version.

Why "Similar" Is Harder Than "Identical"

An exact duplicate shares the same byte-for-byte content. A near-duplicate does not. Two photos can look nearly identical yet differ in:

• Timestamp — burst shots taken milliseconds apart
• File size — a HEIC original versus its JPEG export
• Dimensions — a full-resolution master and a resized share copy
• Colour profile or metadata — the same image edited in Lightroom, saved as a new file
• Slight framing — you moved the camera a hair between two shots

Because these files differ in bytes, simple checksum-based tools miss them entirely. Catching near-duplicates requires perceptual comparison — analysing visual content rather than raw data.

How Similarity Detection Works on Mac

Perceptual hashing is the technique behind most near-duplicate finders. The algorithm scales each image to a tiny thumbnail, converts it to greyscale, and reduces it to a compact hash (often 64 bits). Two photos with similar content produce hashes that differ in only a few bits. The Hamming distance between hashes — the number of differing bits — becomes a similarity score. A distance of 0 means identical content; a distance of 4–10 typically means "looks the same to a human eye."

macOS itself does not expose a built-in perceptual hash tool, but the Photos app has its own similarity logic, and the command line has options for more technical users.

Method 1 — Photos App (Built-in, No Extra Software)

Apple added a Duplicates album to Photos in macOS Ventura (13) and later. It detects visually similar images, not just byte-identical ones.

1. Open Photos.
2. In the left sidebar, scroll to Utilities and click Duplicates. If you don't see it, give Photos a few minutes after launch — it scans in the background.
3. Photos groups similar shots together and marks one as the suggested keep (usually the higher resolution or sharper version).
4. Click Merge to keep the best copy and move the others to Recently Deleted, or click each group individually to inspect before deciding.
5. When done, empty Recently Deleted inside Photos to free disk space permanently.

Limitations: This only works for images already in your Photos library. Photos stored in Finder, on an external drive, or in a different library are not scanned.

Method 2 — Finder and Terminal for Photos Outside the Library

If you keep photos in plain folders — a common pattern for photographers working with RAW files in Capture One or DxO PhotoLab — you need a different approach.

Find burst shots by timestamp

Burst photos taken with an iPhone share a common filename prefix (IMG_XXXX_BURST) and cluster within a tight time window. You can list files by creation time to surface them:

ls -lt ~/Pictures/Imports/ | head -60

This shows the 60 most recently created files, sorted newest-first. Burst shots appear as a consecutive block.

Find large groups of near-identical filenames

Camera files often increment numerically. A long run of sequential names (IMG_4020 through IMG_4031) is a strong signal of a burst or repeated attempts at the same scene:

find ~/Pictures -name "*.jpg" -o -name "*.heic" | sort | uniq -D

That command won't catch visually similar files with unrelated names, but it is a fast first pass for burst sequences.

Find RAW + JPEG pairs

Many cameras write both a RAW file and an embedded JPEG. If you've imported both, you likely have pairs with matching basenames but different extensions:

find ~/Pictures -type f \( -iname "*.cr3" -o -iname "*.arw" -o -iname "*.nef" \) | while read raw; do
  base="${raw%.*}"
  if [ -f "${base}.jpg" ] || [ -f "${base}.JPG" ]; then
    echo "Pair: $raw"
  fi
done

Review each pair and decide whether you need the RAW, the JPEG, or both before deleting anything. Deleting a RAW master is permanent and irreversible.

Method 3 — Crumb's Similar-Photo Grouping

For a visual, point-and-click workflow that covers photos outside the Photos library, Crumb includes a Duplicates scanner that groups near-duplicate images side by side so you can see which shot is sharper, better exposed, or higher resolution before deciding what to remove.

Unlike the Photos app, Crumb scans any folder on your Mac — your Downloads folder full of re-exported screenshots, an external drive of DSLR imports, or a project folder in ~/Documents. After scanning, it shows grouped photos with file size, resolution, and a keep/remove selector per group. You choose which copy to keep; Crumb moves the discards to Trash so you can review before they are permanently gone.

If you are unsure whether a folder is safe to clear, Crumb's built-in AI can explain what a folder is for and flag whether removing files from it carries any risk — useful when you encounter unfamiliar cache or thumbnail directories alongside your photo libraries.

You can download Crumb to try the Duplicates scan; the free tier allows one full cleanup run.

How to Pick the Best Shot from a Group

Once you have identified a group of similar photos, use these criteria to decide which to keep:

In a burst sequence, the first or last frame is often the sharpest (the camera's autofocus settles or the subject's expression peaks at a specific moment). There is no universal rule — look at the actual pixels.

What Is Safe to Delete and What Is Not

Before removing anything, understand the risk level:

• Safe: Redundant burst frames you have reviewed and rejected. JPEG exports you created from a RAW master you are keeping. Screenshots you have already filed or used.
• Caution: The "lower resolution" copy in a pair — verify it is genuinely a derivative and not the only copy of that moment.
• Do not delete without backup: Any photo for which there is no other copy, even if the quality is mediocre. Storage is cheap; lost moments are not.
• Never delete blindly: Do not run any automated bulk-delete operation without reviewing the list first. Always send files to Trash rather than permanent deletion so you have a recovery window.

Keeping Your Photo Library Tidy Long-Term

A few habits prevent similar-photo clutter from accumulating:

• Review bursts immediately after importing, while you still remember which shot worked.
• Set your camera or iPhone to save only the top burst pick automatically (Settings > Camera > Keep Normal Photo, toggle off to save only the best burst frame).
• Choose one canonical location — Photos library or a dedicated folder hierarchy — not both.
• Run a similarity scan every few months rather than letting years of near-duplicates accumulate.

Conclusion

Finding similar photos on Mac requires a different approach from simple duplicate detection — perceptual comparison, manual review of burst groups, and a clear decision framework for choosing the best version. The built-in Photos Duplicates album covers images in your library; Terminal commands and tools like Crumb extend that coverage to photos living anywhere on your disk. Take a few minutes to review each group before deleting, and your photo collection will be smaller, more navigable, and genuinely better.