Image Colour Picker

What the extractor is doing

A photograph can contain hundreds of thousands of distinct colours, so the first job is to throw most of them away. The image is drawn onto a small canvas — around 140 pixels on the long edge — and each visible pixel is assigned to a coarse bucket based on its red, green and blue values. Buckets are then ranked by how many pixels landed in them, and the winners are averaged back into a single representative colour.

Finally, any candidate that is too close to one already chosen is discarded. Without that step you would get eight near-identical shades of the same blue sky, which is accurate but useless. The threshold is deliberately generous, so the eight results are visibly different from one another.

Reading the result honestly

The palette tells you what the image is made of, not what it feels like. A photo of a forest will return a lot of dark greens and very little else, even though the impression it gives is cool and calm rather than green. That gap between composition and impression is useful to know about before you build a whole brand on an extracted palette.

Two adjustments usually help. First, ignore any colour that appears with less than about five per cent of the pixels — it is probably a stray highlight. Second, take the extracted set into the palette generator and look at what the harmonies do with your favourite colour from it; the extracted palette is a starting point, not a finished scheme.

Practical uses

• Matching a website to existing photography. Extract the colours from the hero image, then pick one as the accent so the page and the picture agree.
• Planning an interior. Photograph the room in daylight and see which tones actually dominate, rather than which ones you remember.
• Building a mood board. Run several images through and keep the colours that recur — that overlap is the real palette.
• Checking a design against reality. If your brand blue never appears in the product photography, the two will always look slightly disconnected.

Why nothing is uploaded

Most online image tools send your file to a server, process it there and send back a result. That means your image exists on someone else’s disk, however briefly, and you have to trust their retention policy. This tool does not need a server: the browser can decode an image and read its pixels directly, and it has been able to do so for years.

The practical consequence is that you can safely use it with client work, unpublished product shots or personal photographs. It also works offline, and it is fast, because there is no round trip.