color.js

Extract colors from an image. Works in the browser and in hybrid environments like Electron. Supports images that are present in the DOM, as well as (external) URL's. Size: ~0.75 KB (min + gzip).

Installation

npm install color.js

Or directly in the browser:

<script src="https://unpkg.com/color.js@1.2.0/dist/color.js"></script>

Usage example:

import { prominent } from 'color.js'

prominent('js-logo.jpg', { amount: 1 }).then(color => {
  console.log(color) // [241, 221, 63]
})

// Or with different syntax:

const color = await prominent('js-logo.jpg', { amount: 1 })
console.log(color) // [241, 221, 63]

When used directly in the browser:

<script src="https://unpkg.com/color.js@1.2.0/dist/color.js"></script>

colorjs.prominent('js-logo.jpg', { amount: 1 }).then(color => {
  console.log(color) // [241, 221, 63]
})

API

Prominent

Returns the most used color(s) in an image. Can be requested as a single color or palette of colors (see amount).

import { prominent } from 'color.js'

prominent('img.jpg').then(colors => ...)

Average

Returns the average color of an image.

import { average } from 'color.js'

average('img.jpg').then(color => ...)

Options

You can pass two arguments, an image and a configuration object.

Image (required)

Can be a URL or DOM element.

average('img.jpg')

const img = document.getElementById('photo')
average(img)

average('https://example.com/image.jpg')

When using an external image, CORS should of course be enabled on the source.

Configuration (optional)

Amount

Default: 3

Only applicable for prominent.

The amount of colors that should be returned. When set to 1 a singular value is returned, otherwise an array of values.

prominent('img.jpg', { amount: 5 })

Format

Default: 'array'

The format in which colors should be returned. Options are 'array' and 'hex'.

[241, 221, 63] // 'array'
'#f1dd3f' // 'hex'

average('img.jpg', { format: 'hex' })

Group

Default: 20

Configures how many similar colors should be combined into one color. A value of 1 would mean every individual color would be considered, but this is often not ideal. Especially in photographs there's usually a lot of color data, and grouping colors could give more usable results. In the first example below, group is set to 5 and a lot of individual colors in the sea are returned. When more grouping is applied (30 in the second example), the results become more distinct.

prominent('img.jpg', { group: 30 })

Sample

Default: 10

Configures how many pixels of an image should be processed. For example, a value of 20 means every 20th pixel is interpreted. A higher value means less accurate results, but better performance. An example of default sampling on an image:

average('img.jpg', { sample: 10 })

Browser support

Pretty much everything (> 0.2%), except for IE.