An image color extractor on Windows identifies the most frequent visible colors in a local image and returns a ranked palette with HEX, RGB, and sampled-area percentages.
The Image Color Extractor handles this palette-style task in the current browser. Choose a supported image up to 20 MB and a palette size of 3, 5, 8, or 12; the browser analyzes a scaled local copy and groups nearby RGB pixels. The file is not posted to Lizely or a remote image service.

Extract an image color palette on Windows
- Open the extractor page in your browser on the Windows computer. Have the image you want to analyze available as a local file.
- Choose the image. The input limit is 20 MB. The interface accepts common PNG, JPEG, WebP, GIF, BMP, and AVIF types, although successful decoding ultimately depends on the browser.
- Select a palette size. Choose 3, 5, 8, or 12 colors according to how much detail you need from the image.
- Wait for the local canvas analysis. The browser decodes the image, scales it when necessary, groups similar RGB pixels, and ranks the groups by their sampled pixel counts.
- Compare the swatches with the image. Look for the regions that occupy the most visible area. If two important regions have merged or a meaningful accent is missing, return to the palette setting and request more colors.
- Select a swatch to copy its HEX code. Copy the value into your design draft, then check any foreground and background pairing with Color Contrast Checker before using it for text. For exact brand requirements, inspect the intended source pixels in a color-managed design application.
Each result includes a HEX code, an RGB triplet, and the percentage of sampled visible pixels assigned to that color group. These values describe the analyzed image rather than an official brand palette or a guarantee that every source pixel exactly matches the displayed swatch.
Choose a palette size that fits the job
The palette setting determines how many top-ranked groups you can review. It does not change the underlying analysis or turn a grouped result into an exact pixel measurement.
| Requested colors | Practical purpose |
|---|---|
| 3 | A short visual summary of the image’s dominant color groups |
| 5 | A broader palette for comparing main and secondary groups |
| 8 | A more detailed set of frequent groups |
| 12 | The largest palette offered for investigating complex images or merged groups |
Start with the smallest useful palette, then compare the results with the visible composition. A three-color summary may be enough for a simple mood board, while an eight- or twelve-color palette can expose additional groups in a detailed illustration. The percentage beside each swatch describes sampled area, not visual importance: a large background may rank first simply because it covers much of the image.
How the browser builds the palette
- Decoding and scaling: The browser decodes the local image and proportionally scales it down only when necessary so neither dimension exceeds 512 pixels.
- Transparency handling: Pixels with an alpha value below 128 are ignored, preventing mostly transparent backgrounds from dominating the results.
- Color grouping: Each visible red, green, and blue channel is placed into one of eight coarse ranges. The three channels form at most 512 RGB buckets.
- Averaging and ranking: The tool counts the pixels in each occupied bucket, averages the original channel values within that bucket, and sorts the buckets by descending count. Ties use a stable numeric bucket order.
- Percentage calculation: Each displayed percentage is based on the sampled visible pixels assigned to that group, not on the file’s original pixel count.
This method is transparent and deterministic, but it is not a perceptual clustering model. Two colors that look very similar can fall into adjacent buckets, while several subtly different shades can merge into one averaged swatch. Brightness, gamma, device color profiles, and human visual sensitivity are not modeled.
Read the swatches and percentages correctly
The HEX value provides a compact color code suitable for a design draft, while the RGB triplet shows the red, green, and blue channel values behind the result. The percentage helps explain why a group appears near the top of the palette, but it cannot tell you whether the color is attractive, suitable for text, or important to the image’s meaning.
A swatch can represent an average of pixels within its bucket rather than one perfectly represented source color. It is therefore best treated as a starting point for exploration. If a thin line, tiny icon, or other small accent is absent, that does not mean the feature is visually unimportant; it may have disappeared during scaling or ranked below the selected palette size.
Palette extraction or exact pixel picking?
The right choice depends on the question you need answered. Use extraction when you want to know which broad colors occupy the most sampled area. Use the Image Color Picker when the color of one button, logo edge, or sampled detail is the priority.
| Feature | Image Color Extractor | Image Color Picker |
|---|---|---|
| Question answered | Which broad color groups occur most often across the image? | What is the color at the location I select? |
| Input | A complete local image | A chosen point in an image |
| Result | Ranked grouped HEX, RGB, and percentage values | The exact RGB value at the selected point |
| Typical use | Mood boards, themes, CSS-variable starting points, chart colors, and illustration studies | Precise interface details, edges, and sampled design elements |
These tools answer different questions and can be used together. An extracted palette can establish the image’s main color direction, while a pixel sample can confirm the value of a specific element that was too small or too subtle to appear in the ranked results.
Turn the palette into design starting points
- Presentations and mood boards: Begin with three colors for a broad visual direction, then request five or more when you need supporting shades.
- CSS variables: Copy selected HEX values into a draft variable list, but run text and background combinations through Color Contrast Checker before publishing them.
- Charts and diagrams: Use the ranked colors as candidates, while checking that important categories are visually distinguishable; frequency ranking does not test accessibility.
- Illustration and logo studies: Review the dominant groups for exploration, but inspect the original file in a design application when an exact brand color matters.
Keep the extracted values in a working draft rather than treating the palette as a final specification. The results are most useful for quickly narrowing down colors that deserve closer inspection.
Privacy, file limits, and errors on Windows
The browser represents the selected file with a temporary Object URL and draws the decoded image to an in-memory canvas. It does not post the file to Lizely or a remote image service. Replacing the image or leaving the page releases the temporary URL.
- File size: Images can be selected up to 20 MB. The pixel-processing function also has a separate defensive buffer limit, so a file that meets the 20 MB limit can still encounter the second limit during decoded-pixel processing.
- Format support: The interface accepts PNG, JPEG, WebP, GIF, BMP, and AVIF, but actual decoding depends on the browser used on the Windows PC.
- No visible pixels: A file with no sufficiently visible pixels returns a clear error rather than a meaningful palette.
- Animated images: Analysis uses the frame that the browser decodes for the canvas, not a combination of every animation frame.
- Small visual details: Bounded scaling can reduce their contribution, while area-based ranking can place them below the requested number of colors.
Know when the result is only exploratory
The extractor uses coarse sRGB channel buckets and does not model ICC profiles, perceptual distance, Pantone inks, or print conditions. Its palette should not be treated as an official brand palette, a print ink formula, trademark permission, or proof that a text and background combination is accessible.
When exact brand measurement matters, open the original file in a color-managed design application and sample the intended source pixels. When preparing text colors, run each foreground and background pair through Color Contrast Checker. For point-specific work, sample the exact location separately; for broad palette exploration, compare the ranked swatches with the image and adjust the palette size when needed.
If you're weighing options, Invert Image Colors for Instagram: Local PNG Workflow covers this in detail.
If you're weighing options, Image Color Picker on Mac: Get Exact HEX and RGB covers this in detail.