How do I get the exact color value of one pixel in an image?

Move to the target spot and the loupe shows the surrounding 11×11 pixels with the center cell outlined; clicking reads that pixel's 8-bit sRGB value, 0–255 per channel. For finer control, focus the image and move with the arrow keys one pixel at a time, or 10 pixels with Shift, then press Enter to record the color under the cursor.

How are the dominant color percentages calculated?

The image is first downsampled to about 40,000 pixels, pixels below 50% opacity are ignored, and median cut splits the remaining colors into 4–12 groups. Each share equals the pixels in a group divided by all counted pixels, so the shares add up to 100%. The algorithm has no random steps, which means the same image always produces exactly the same palette.

Why do picked colors differ from my design software?

The usual cause is color management. The browser converts embedded profiles such as Display P3 or Adobe RGB to sRGB before reading, while design software may show values in the original profile. A 16-bit PNG is also reduced to 8 bits per channel. On top of that, JPEG compression makes neighboring pixels differ by a few levels, so pick from a lossless 8-bit sRGB original for exact matches.

Image color picker

Click an image to read pixel colors as HEX, RGB and HSL, and extract dominant colors with their share

Computed in your browser

Load an image, move over it to see the loupe and click to record colors; dominant colors and shares appear below automatically.

Choose or drop an image You can also paste an image from the clipboard. Images are processed in this browser and never uploaded. One JPG, PNG, WebP, AVIF, BMP or GIF image at a time, up to 50 MB.

Example

Input

sample.png (generated, 1200×800) · 6 dominant colors · pick at the center (600, 400)

Example result

HEX/RGB/HSL of the center pixel, 6 dominant colors sorted by share, and the average color

The sample has fine noise, so neighboring pixels differ by a few levels; use the loupe to confirm the exact pixel.

Technical details & limits

Colors are read from Canvas 8-bit sRGB pixels (0–255 per channel), after the browser converts any embedded ICC profile to sRGB. Images longer than 4,096 px are scaled down for picking, with coordinates mapped back to original pixels. Palette extraction downsamples to about 40,000 pixels, ignores pixels below 50% opacity and splits colors into the requested number of groups by median cut; share equals pixels in a group divided by counted pixels. The algorithm has no random steps, so the same image always gives the same result.

About this tool

Load an image and click anywhere to read that pixel's color. A loupe shows the surrounding 11×11 pixels for precise placement, and arrow keys move pixel by pixel with Enter to record. Picked colors are listed as HEX, RGB and HSL with copy buttons, keeping up to 24 entries. The tool also extracts 4–12 dominant colors with their share and reports the average color of the whole image. Images are read in the browser and never uploaded.

How to use

Drop, choose or paste an image.
Move over the image to see the loupe; click, or position with arrow keys and press Enter, to record a color.
Copy the values you need, or choose how many dominant colors to extract.

Input & output

Input: One JPEG, PNG, WebP, AVIF, BMP or GIF (first frame) image, up to 50 MB and 50 megapixels.
Output: HEX, RGB and HSL values, pixel coordinates, 4–12 dominant colors with shares and the average color.
Limits: Values are display 8-bit sRGB; no CMYK, Lab or print color numbers; images over 4,096 px are scaled down for picking.

Notes

JPEG compression causes small differences between neighboring pixels; pick from a lossless original such as PNG for exact values. Convert results further with the color converter.

FAQ

How do I get the exact color value of one pixel in an image?: Move to the target spot and the loupe shows the surrounding 11×11 pixels with the center cell outlined; clicking reads that pixel's 8-bit sRGB value, 0–255 per channel. For finer control, focus the image and move with the arrow keys one pixel at a time, or 10 pixels with Shift, then press Enter to record the color under the cursor.
How are the dominant color percentages calculated?: The image is first downsampled to about 40,000 pixels, pixels below 50% opacity are ignored, and median cut splits the remaining colors into 4–12 groups. Each share equals the pixels in a group divided by all counted pixels, so the shares add up to 100%. The algorithm has no random steps, which means the same image always produces exactly the same palette.
Why do picked colors differ from my design software?: The usual cause is color management. The browser converts embedded profiles such as Display P3 or Adobe RGB to sRGB before reading, while design software may show values in the original profile. A 16-bit PNG is also reduced to 8 bits per channel. On top of that, JPEG compression makes neighboring pixels differ by a few levels, so pick from a lossless 8-bit sRGB original for exact matches.