reverse image search July 6, 2026 ⏱ 6 min read

Reverse Image Search: Upload One Photo, Find Every Copy

Upload a reference image or video and Sakarto finds every similar copy in a folder. 7 algorithms explained—Color Signature, aHash, BlockHash, dHash, pHash, wHash, and ORB. Reference pinned, protected, and ready to compare.

Finding duplicate images normally means scanning a whole folder and grouping everything visually similar. That’s useful when you’re cleaning up a messy photo library and you don’t know what’s in there. But sometimes you know exactly which file you want copies of—a specific product photo, a company logo, or a photo of a person you need to find in a large archive.

That’s where reverse image search comes in. Instead of grouping everything, you upload one reference file first. Sakarto processes it immediately, then scans your chosen folder and returns only the files that look like that specific reference. The reference is pinned at the top with a protected badge—it can never be accidentally moved or deleted.

It’s like Google reverse image search, but entirely in your browser, with zero uploads, and with seven different algorithms to choose from.

Quick reference

Algorithm	Best for	Threshold direction	Speed
Color Signature	Same colour palette, same scene	Lower = stricter	Fast
aHash	Large folders, speed priority	Lower = stricter	Fastest
BlockHash	JPEG-compressed copies	Lower = stricter	Very Fast
dHash	Brightness/exposure-adjusted	Lower = stricter	Very Fast
pHash	All-round reliability	Lower = stricter	Fast
wHash	Speed + quality balance	Lower = stricter	Fast
ORB	Rotated, cropped, perspective	Higher = stricter (inverted)	Slower

How reverse image search is different from find duplicates

The standard Find Duplicate Images & Videos workflow scans a folder and compares every file against every other file. That’s O(n²) complexity—fine for a few hundred images, but it gets slower as the folder grows.

Reverse Image Search does something much smarter. It hashes your reference once, then compares each folder file only against that reference. That’s O(n) complexity—linear scaling. A folder of 10,000 images takes roughly the same time per file as a folder of 100. The scan speed depends only on the number of files, not how many of them are similar to each other.

This makes reverse search the faster choice for large folders where you already know what you’re looking for.

The reference protection system

The reference file is treated differently from every other file in the scan. When you upload it, Sakarto immediately:

Computes its fingerprint using your chosen algorithm
Shows a thumbnail preview so you can confirm the right file was loaded
Pins it at the top of the results as soon as scanning starts
Displays a blue REFERENCE badge on the card
Prevents it from being moved or deleted—Move and Delete buttons are disabled on the reference card

This last point is critical. If you’re using reverse search to find copies of a file you want to keep, you don’t want to accidentally delete the original.

The seven algorithms, applied to reverse search

For detailed explanations of how each algorithm works, see Seven Ways to Catch a Duplicate — this guide focuses on how they behave specifically in reverse search mode.

Color Signature — same colour palette, same scene

The only algorithm that actually compares colour—not just brightness. Finds copies that share the same colour palette as your reference. Will not match colour-shifted versions.

Open Color Signature Reverse Search

aHash — speed for large folders

The simplest and fastest algorithm. Fast Mode (reading JPEG EXIF thumbnails) makes this even faster—5–10× speedup with near-identical accuracy.

Open aHash Reverse Search

BlockHash — JPEG-compressed copies

Block averaging absorbs JPEG compression noise. Finds heavily re-compressed versions of your reference that pixel-level methods would miss.

Open BlockHash Reverse Search

dHash — brightness and exposure-corrected copies

Compares the direction of brightness change—not absolute values. Robust to exposure shifts and brightness corrections.

Open dHash Reverse Search

pHash — the all-rounder

Uses the Discrete Cosine Transform to extract low-frequency “essence.” The most reliable general-purpose choice for edited copies.

Open pHash Reverse Search

wHash — locally edited copies

Uses the Haar Wavelet Transform on a 64×64 canvas. Handles local edits better than pHash because the multi-scale decomposition captures unchanged regions.

Open wHash Reverse Search

ORB — rotation, cropping, and perspective

The only algorithm that handles rotation, cropping, and perspective distortion. Uses OpenCV.js (WebAssembly) to detect keypoints. Higher threshold = stricter (inverted). Slower but uniquely capable.

Open ORB Reverse Search

Reference-first workflow in action

Open the reverse search page for your chosen algorithm.
Upload your reference from anywhere on your computer.
Click “Choose Folder to Search” and pick the folder where copies live.
Matches appear live as the scan runs—you don’t need to wait for the full scan to finish.
Adjust the threshold to tighten or loosen—results re-filter instantly.
Preview and compare using the built-in media viewer and side-by-side compare modal.
Select matched files and move or delete them. The reference card is always excluded from operations.

Privacy and local execution

Like every other Sakarto tool, reverse image search runs entirely in your browser. Open DevTools → Network tab during any scan: zero outbound requests. No uploads, no telemetry, no data leaving your device.

The verdict

Reverse image search is faster for large folders, protects your reference file, and gives you a clear, focused view of exactly what you’re looking for. With seven algorithms to choose from, you can match the algorithm to the kind of duplicates you expect to find.