deno-imagehash

Perceptual JPEG image hashing for Deno via Blockhash.js

Usage

import { imageHash, hammingDistance, DecodedImage } from "https://deno.land/x/imagehash/mod.ts";

serve(async (request: Request) => {
    const rawImageData: Uint8Array = request.body
    const hash: String = await imageHash(rawImageData, 10);
    
    // OR
    
    const decodedImage: DecodedImage = jpeg.decode(rawImageData)  // decodedImage = { width: 100, height: 100, data: ... }
    const hash: String = await imageHash(decodedImage, 10);
    
    // Hamming Distance
    const bitDifference = hammingDistance(hash1, hash2); // returns number
    
    return new Response(JSON.stringify({hash, bitDifference}), {status: 200})

API

imageHash(imgData, bits?)

Returns: string

Returns the hash value as a string of hexadecimal characters.

imgData

Type: Uint8Array or DecodedImage/jpeg.Image

The image data to hash. This can be a Uint8Array or an object that matches DecodedImage: {data: `Uint8Array`, width: `number`, height: `number`})

bits = 10

(Optional)

Type: number / Default: 10

The number of bits to use for the hash. The higher the number, the more accurate the hash, but the longer it will take to compute.

hammingDistance(hash1, hash2)

Calculate distance in bits between two hashes.

Credits

This module uses the following projects with some changes to work with Deno:

• Blockhash.js

• jpeg.ts

Inspired by: deno-image