palette-extract-rs

A Rust lib for extracting a color palette from an image!

Based on a port of Leptonica's modified median cut quantization algorithm.

Example

¹

Credits

This Rust implementation of modified median cut quantization (MMCQ) is adapted from Kazuki Ohara's ColorThiefSwift.

Special thanks

• Sven Woltmann - for the Java Implementation. ColorThiefSwift is a port of this.
• Lokesh Dhakar - for the JavaScript version of Color Thief which inspired many of the ports.
• And of course, Dan Bloomberg for the original paper on MMCQ, along with the reference implementation found in Leptonica.

Installation

To use, add the following to Cargo.toml under [dependencies]:

palette_extract = "0.1.0"

Usage

Using the library consists of calling get_palette_rgb or get_palette_with_options with a set of RGB or RGBA pixels represented as a u8 slice.

Basic

A minimal example using 4 red pixels represented in RGB looks like this:

use palette_extract::{get_palette_rgb, Color};

fn main() {
    let pixels: [u8; 12] = [255, 0, 0, 255, 0, 0, 255, 0, 0, 255, 0, 0];

    let r = get_palette_rgb(&pixels);

    assert_eq!(r.len(), 1);
    assert_eq!(r[0], Color::new(252, 4, 4));
}

Image from file/somewhere else

Here's one way to extract the color palette of an image by leveraging the image crate to read and decode an image file (see full working example in examples directory):

    let image_path = "./path/to/image.jpg";

    // open and decode the image using the `image` crate
    let img = image::open(image_path).unwrap();

    // grab a reference to the underlying pixels/RGB buffer
    let pixels = img.as_bytes();

    // extract the color palette
    let palette = get_palette_rgb(&pixels);

    // output the extracted color palette
    palette.iter().for_each(|x| println!("{:?}", x));

More usage examples can be found in the examples directory!

Background/Further reading

• The paper describing the MMCQ algorithm: http://leptonica.org/papers/mediancut.pdf

• A great post on different ways to extract color palettes: Color quantization

Footnotes

1. Image credit: João Pacheco ↩