A simple way to get the dominant color of an image in python

Overview

In this post, we’ll explore how to extract the dominant color from an object in an image using three open-source Python libraries. We’ll guide you through the following steps:

1. Remove the background of the image, isolating the object of interest.
2. Extract the dominant color code from the object.
3. Retrieve the color name corresponding to the extracted color code.

Execution

1. Setting Up the Environment

First, let’s create a virtual environment to isolate our dependencies. We’ll use Python 3.9 for this example:

pyenv virtualenv 3.9 dominant-color

2. Removing the Background

To begin, we’ll use the rembg library to remove the background from the image.

Install the rembg package:

pip install rembg

Remove the background from the image:

from rembg import remove

input_path = 'input.png'
output_path = 'output.png'

with open(input_path, 'rb') as i:
    with open(output_path, 'wb') as o:
        input = i.read()
        output = remove(input)
        o.write(output)

You can also run this process directly from the terminal. For more details, visit the rembg repository.

3. Retrieving the Dominant Color Code

Next, we’ll use the colorthief library to extract the dominant color code from the object.

Install the colorthief package:

pip install colorthief

Get the dominant color code:

from colorthief import ColorThief

color_thief = ColorThief('/path/to/imagefile')
# get the dominant color
dominant_color = color_thief.get_color(quality=1)

The colorthief library can also generate a color palette. For more details, visit the ColorThief project page.

4. Retrieving the Color Name

Finally, we’ll use the colornamer library to retrieve the color name corresponding to the dominant color code.

Install the colornamer package:

pip install colornamer

Get the color name:

from colornamer import get_color_from_rgb

get_color_from_rgb([5, 135, 210])

For more information on how to interpret the results, visit the colornamer repository.

Acknowledgments

This post was inspired by the excellent work found in the rembg, ColorThief, and colornamer repositories.