https://github.com/trflorian/auto-color-filter

Demo Applications using different Color Spaces
https://github.com/trflorian/auto-color-filter

color-spaces computer-vision opencv-python python

Last synced: 12 months ago
JSON representation

Demo Applications using different Color Spaces

• Host: GitHub
• URL: https://github.com/trflorian/auto-color-filter
• Owner: trflorian
• License: mit
• Created: 2024-12-25T01:09:04.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2025-05-04T16:55:24.000Z (about 1 year ago)
• Last Synced: 2025-06-03T18:28:45.295Z (about 1 year ago)
• Topics: color-spaces, computer-vision, opencv-python, python
• Language: Python
• Homepage: https://towardsdatascience.com/from-rgb-to-hsv-and-back-again/
• Size: 1.65 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# Color Spaces

![Python](https://img.shields.io/badge/python-3.13-blue.svg)

![GitHub License](https://img.shields.io/github/license/trflorian/auto-color-filter)

Undestaning different color spaces is crucial for Computer Vision projects. In this project I explore different examples where switching between color spaces is extremely useful.

![color_transition_rgb_vs_hsv](https://github.com/user-attachments/assets/ba15e0c7-10e8-413c-9cc5-2bad189e6e86)

## Prerequisites

- [uv](https://docs.astral.sh/uv/)

## Quickstart

You can run the `main.py` file for the main application:

```bash

uv run src/auto_color_filter/main.py

```

By clicking and holding the left mouse button you can add pixels from the image to an inclusion list. Based on this list, a color range will be defined and based on that a mask image is generated.

![auto_filter_demo](https://github.com/user-attachments/assets/9e654688-39d4-4f95-935b-3d36134cc9a4)

## More Demos

### Manual Segmentation

The example provided in `segmentation.py` creates a mask image based on a static HSV range defined in the arguments.

```bash

uv run src/auto_color_filter/segmentation.py

```

### Color Transition

In this demo we compare two different ways to interpolate between two colors:

- Directly with RGB values interpolated -> changes in brightness and saturation even if start and end color have the same brightness/saturation

- Interpolate in HSV color space -> constant brightness/saturation if start and end values match or linear interpolation in case of difference between start and end color

```bash

uv run src/auto_color_filter/segmentation.py

```

## Tests

Using pytest you can run the test suite:

```bash

uv run pytest

```