About This Tool
The palette extractor groups similar image colors into a small set of representative colors. Use it to explore palettes from photos, illustrations, or brand artwork.
How to Use
- Choose or drop in an image to see the default five-color palette.
- Select 3, 5, 8, or 10 colors as needed.
- Review each swatch, color code, and estimated share.
- Copy individual colors, or copy the full HEX list or CSS variables at once.
Supported Formats
- PNG, JPG/JPEG, and static WebP images are supported. Choose or drop in one image at a time.
- Files can be up to 100 MiB, 40 million pixels, and 16384 pixels on either side. Large images need more memory, so choose a size that works well on your device.
Understanding the Results
- Colors are sorted by estimated weight. Each card shows HEX, RGB, and HSL text with copy buttons.
- The selected number is the target palette size. A solid-color image or one with fewer distinct colors may return fewer colors.
- Fully transparent pixels are ignored, and other pixels are weighted by alpha. Black and white are included.
How Colors Are Extracted
- Consistent median-cut grouping brings similar colors together and finds a representative color for each group, using the same method as dominant color extraction. The same image and settings give the same result.
- Dominant color and palette analysis use up to 250,000 pixels, with a maximum side length of 1000 pixels. Large images are scaled down, and the results show the actual analysis size. Small images use all their pixels.
- Three colors summarize the overall palette; ten preserve more variation. Increasing the count does not guarantee that every tiny color detail will appear.
Privacy
Your images are processed only in your browser and are never uploaded to a server.
The tool does not automatically save your input or results. Copy or download anything you need before clearing or leaving the page, and keep the original image.
FAQ
How is a color palette extracted?
The tool reads pixels, groups similar colors into representative clusters, and lists them by weight. Large images are scaled down first, with the analysis dimensions shown in the results.
Why can one image produce several colors?
An image usually contains different colors in its background, subject, shadows, and highlights. A palette summarizes these with several representative colors instead of only one dominant color.
What is the difference between 3 and 10 colors?
Three colors give a concise summary, while ten preserve more color variation. Images with few colors may return fewer than the number selected.
How are transparent pixels handled?
Fully transparent pixels are excluded. Others are weighted by alpha, so semitransparent pixels carry less weight than the same number of opaque pixels.