What Is a Perceptual Hash?
A perceptual hash squeezes an image into a fixed-length fingerprint (64 bits here) while ensuring that visually similar images get similar fingerprints.
It works by extracting low-frequency features in the frequency domain with DCT, ignoring compression, resizing, and slight color shifts that don't affect how we perceive the image.
How to Compute an Image Perceptual Hash
Three steps to compute the pHash of your image:
- Upload a local image (PNG, JPG, WEBP, or BMP supported).
- Click Compute Hash to generate the 64-bit hash value.
- Copy the HEX or binary result for programmatic comparison or archiving.
Tips
- Different compression or resized versions of the same image produce nearly identical hashes.
- A Hamming distance under 10-12 bits indicates similar images.
- Pair this tool with Duplicate Image Finder to deduplicate images in bulk.