search

commonsmachinery / blockhash-js

Image perceptual hash calculation in javascript
http://blockhash.io/
MIT License
172 stars 49 forks source link

readme

blockhash-js

This is a perceptual image hash calculation tool based on algorithm descibed in Block Mean Value Based Image Perceptual Hashing by Bian Yang, Fan Gu and Xiamu Niu.

Installation

This module is installed via npm:

  $ npm install blockhash

Use in the browser

To use this library in the browser, you can build it with Browserify with something like browserify index.js --standalone blockhashjs > blockhash.js

Include it and zlib.js on your page:

<!DOCTYPE html>
<html>
  <head>
    <title>Blockhash</title>
  </head>
  <body>
    <script src="https://github.com/commonsmachinery/blockhash-js/raw/master/node_modules/png-js/zlib.js"></script>
    <script src="https://github.com/commonsmachinery/blockhash-js/raw/master/blockhash.js"></script>
    <script>
      var blockhash = blockhashjs.blockhash;
    </script>
  </body>
</html>

Call blockhash(src, bits, method, callback), where src is an image URL, bits is the number of bits in a row, method is a number 1-2 (see below), and callback is a function with (error, result) signature. On success, result will be array of binary values.

The available methods are:

  1. Quick and crude, non-overlapping blocks
  2. Precise but slower, non-overlapping blocks

Method 2 is recommended as a good tradeoff between speed and good matches on any image size. The quick ones are only advisable when the image width and height are an even multiple of the number of blocks used.

License

Copyright 2014 Commons Machinery http://commonsmachinery.se/

Distributed under an MIT license, please see LICENSE in the top dir.

Contact: dev@commonsmachinery.se