Grayscale Converter
Converted pixel by pixel in your browser. Nothing is uploaded.
Luminance (Rec. 709), perceptually correct
Weights each channel by how bright the eye perceives it. Green dominates because human vision is most sensitive to it.
Choose an image to start.
The greyscale converter offers three weightings rather than one, because there is no single correct way to turn colour into grey. The naive average treats red, green and blue equally, which produces a flat result. Human vision is far more sensitive to green than to blue, and the perceptual weightings account for that.
How it works
grey = R x wr + G x wg + B x wb
- Rec. 709
- 0.2126, 0.7152, 0.0722, the modern HD standard, perceptually correct
- Rec. 601
- 0.299, 0.587, 0.114, the older television standard, still used by many tools
- Average
- 1/3 each: mathematically simple, visually wrong
Green carries roughly 71 percent of perceived brightness under Rec. 709 and blue only 7. Averaging the channels therefore makes blues far too light and greens too dark, which is why an averaged conversion looks washed out and low in contrast.
The strength slider blends between the original and the greyscale result, so partial desaturation is possible without a separate tool.
Examples
A pure blue pixel
Colour
rgb(0, 0, 255)
Result
Rec. 709 gives 18 · Rec. 601 gives 29 · Average gives 85
The average makes pure blue a mid grey, which is nothing like how bright it appears. This single case explains why the naive method looks wrong.
A pure green pixel
Colour
rgb(0, 255, 0)
Result
Rec. 709 gives 182 · Average gives 85
Green should be far brighter than blue in greyscale. The average renders them identically, collapsing a real difference.
Frequently asked questions
Which greyscale method should I use?
Rec. 709 for anything modern. It matches how sRGB displays and human vision actually behave. Rec. 601 if you need to match output from older software or video tooling, which still uses it widely. The average is included to show why it looks wrong, not because it is ever the better choice.
Why does averaging the channels look so flat?
Because it ignores that the eye is about ten times more sensitive to green than to blue. Averaging makes pure blue and pure green the same shade of grey, when perceptually green is far brighter. The result loses contrast that was in the original.
Is the conversion reversible?
No. Three channels are collapsed into one, so the colour information is gone. Keep the original file. Colourising a greyscale image is guesswork, not recovery.
Does the file get smaller?
Not necessarily. The output is still a three-channel PNG with equal values in each, so the dimensions and format decide the size. Saving as a true single-channel greyscale image needs an encoder option this tool does not expose.