# Colour-blindness checker for scientific figures

> Upload a figure or chart and see it with protanopia, deuteranopia, tritanopia and monochrome vision. Finds colours that become confusable. Private, in-browser.

URL: https://www.labcanvas.com/tools/scientific-figure-color-blindness-checker

Upload a figure, chart or micrograph to see how it looks with protanopia, deuteranopia, tritanopia and monochrome vision, side by side. The checker also finds the main colours in the image and lists pairs that become hard to tell apart, so you know what to fix before submission. The image never leaves your browser.

## Quick answer

- **What it does:** Shows an uploaded figure as seen with protanopia, deuteranopia, tritanopia and monochrome vision, and lists colour pairs that become hard to tell apart.
- **Formula:** Machado, Oliveira & Fernandes 2009 dichromacy matrices in linear RGB, Rec. 709 luminance for monochrome; a pair is flagged when ΔE > 20 normally but ΔE < 12 after simulation.
- **Example:** Red #D62728 and green #2CA02C differ by ΔE ≈ 120 for normal vision but only ΔE ≈ 7 with deuteranopia, so the checker flags them as a confusable pair.

Free, in the browser, no account needed. Updated 2026-11-08.

## Features

- Protanopia, deuteranopia, tritanopia, achromatopsia
- Adjustable severity (anomalous trichromacy)
- Automatic list of confusable colour pairs
- Download each simulation as PNG
- Images stay on your device

## How to use it

1. **Upload the figure**: Drop a PNG, JPEG, WebP or SVG of your figure, graph or image.
2. **Compare the simulations**: See all four colour-vision types at once, or one side by side with normal vision. Lower the severity to simulate milder, more common forms.
3. **Fix what collides**: Read the list of colour pairs that look alike and change one of them, or add labels, patterns or marker shapes.

## At a glance

| | |
| --- | --- |
| Input | An image of your figure (it stays on your device) |
| Simulations | Protanopia, deuteranopia, tritanopia and achromatopsia, with adjustable severity |
| Method | Machado, Oliveira and Fernandes (2009) matrices in linear RGB; Rec. 709 luminance for monochrome |
| Flags | Colour pairs with \(\Delta E > 20\) in normal vision but \(\Delta E < 12\) after simulation |
| Output | A PNG of each simulation |

## How it works

Each pixel is converted to linear RGB and multiplied by the **Machado, Oliveira and Fernandes (2009)** dichromacy matrix for protanopia, deuteranopia or tritanopia; lower severities blend towards normal vision. Monochrome uses Rec. 709 luminance. To find problems, the image’s main hues are clustered and every pair is compared in CIELAB: pairs that differ clearly for normal vision (\(\Delta E > 20\)) but are close after simulation (\(\Delta E < 12\)) are reported.

## Questions

### How common is colour blindness?

Red–green colour-vision deficiency affects about 1 in 12 men and 1 in 200 women of northern European ancestry. Blue–yellow (tritan) deficiency and complete colour blindness are rare.

### Which colours are hardest for colour-blind readers?

Red versus green, and red/green versus brown or orange of similar lightness, are the classic problems for protan and deutan vision. Blue versus green and yellow versus pink can be hard for tritan vision.

### Do journals require colour-blind-friendly figures?

Many journals, Nature among them, ask authors to avoid red–green combinations or otherwise make figures readable for colour-blind readers. Check your journal’s figure guidelines.

### Is my image uploaded?

No. The simulation runs in your browser with JavaScript; the file is not sent to a server.

### What does severity mean?

At 1.0 a cone type is missing (dichromacy). Lower values approximate anomalous trichromacy, where the cone is present but shifted; it is more common and milder.