About Colorblind Lab
Colorblind Lab is a color vision self-check and a color blindness simulator that run right in your browser. The check plates aren't pre-drawn images — they're computed fresh every time you open the page. That means nobody can pass by memorizing plates, and with a seed value you can retake the exact same plates.
The result only estimates type and severity from how many red-axis (protan), green-axis (deutan), and blue-yellow-axis (tritan) plates you read — it does not diagnose. The result screen also reminds you that results can shift with your screen and lighting.
This result is not a medical diagnosis. It is an estimate for reference that can vary widely with your screen's color reproduction, brightness, night mode (blue light filter), ambient lighting, and viewing distance. To confirm the exact type and severity, see an eye doctor for standard tests such as the anomaloscope.
Key features
- Procedurally generated dot plates — about 1,400 dots of varying size fill a circle, with the number and background painted in colors from opposite sides of a confusion line. Each dot's brightness is randomized so the number can't be read by brightness differences.
- 12-, 16-, and 20-plate modes — 1–2 control plates are mixed in with red, green, and blue-yellow axis plates, ranging from large color differences to small ones.
- Type and severity estimate — misses per axis sort the result into normal range, mild, moderate, or strong, and the gap between red-axis and green-axis misses shows a protan or deutan leaning. Control plate misses and inconsistent answers trigger separate warnings.
- Plate review and recent results — see the correct answer and your answer for every plate, with your 20 most recent results saved in this browser.
- Color blindness simulator — convert photos or design mockups to protan, deutan, or tritan (severity 0–100%) or achromatopsia, and save them as PNG.
- Types and guides — eight types of color vision deficiency, inheritance, daily life and careers, colorblind-friendly palettes, and test methods, all with sources.
How to use it
- Turn up screen brightness and switch off night mode, blue light filters, and color filters, then choose the number of plates (12, 16, or 20).
- Press “Start check” and enter the number you see on each plate using the keypad (or your keyboard). If you can't see one, press “Can't see.”
- After the last plate, you'll see correct answers per axis, the estimated result, and a plate-by-plate review. Use “Same plates again” to recheck under different conditions.
- To learn more about a type in your result, tap “Learn about this type”; to see an image through someone else's eyes, head to the simulator.
How the dot plates are made
Plot colors on the CIE 1976 u'v' chromaticity diagram, and the colors a given color vision deficiency can't tell apart fall on straight lines radiating from a single point (the copunctal point). These lines are confusion lines. Colorblind Lab uses the copunctal points of the Cambridge Colour Test — protan (0.678, 0.501), deutan (−1.217, 0.782), tritan (0.257, 0.0) — and spreads the number color and background color apart from each plate's center color along these directions.
The chromaticity distance between number and background runs from 0.100 → 0.019 u'v' units on the red and green axes and 0.110 → 0.036 on the blue-yellow axis. Each dot's brightness (luminance) is picked at random from six levels so that the number can't be revealed by brightness alone rather than color. Every plate's colors are automatically checked to fall within the sRGB gamut.
The digits use a hand-built stroke font so they look identical regardless of device fonts. No designs or colors from the Ishihara plates are used.
Frequently asked questions
Can this check diagnose color blindness or color weakness?
No. Colorblind Lab is an on-screen self-check, so it depends heavily on your monitor's color reproduction and your lighting. The result is an estimate for reference; the exact type and severity must be confirmed with standard tests at an eye clinic, such as the anomaloscope or HRR.
Is this the same as the Ishihara test?
It uses the same family of principles (pseudoisochromatic plates), but the plates are entirely different. Colorblind Lab doesn't use Ishihara designs; it computes new plates every time from confusion lines and luminance noise. So you can't pass by memorizing famous plates.
My results are a little different every time.
The plates are generated fresh each time, and the plates with the smallest color differences sit close to the threshold even for normal color vision, so they can wobble with concentration and screen angle. Missing up to one plate per axis still counts as the normal range. If your results disagree, retake the check with the same seed on another device.
I only missed a lot of blue-yellow (tritan) plates.
Congenital blue-yellow deficiency is rare — fewer than 1 in 10,000 people. The most common cause is a feature that reduces blue light from the screen, such as night mode (Night Shift, Night Light), a blue light filter, or a warm color temperature setting. Turn it off and try again. If it keeps happening, we recommend seeing an eye doctor.
Can women be color-deficient too?
Yes. Red-green color vision deficiency is inherited on the X chromosome, so it's much rarer in women than in men, but it still affects about 0.4–0.5% of women of European ancestry. Blue-yellow deficiency and achromatopsia occur at similar rates in men and women. See the inheritance guide for detailed odds.
Where do images I upload to the simulator go?
Nowhere. Images are read and converted only on a canvas inside your browser and are never sent to a server. Pressing the save button simply downloads the converted result to this device.
Can I take the check on my phone?
Yes. Keep in mind that the smaller the screen, the smaller the dots, and phones are more affected by auto-brightness and viewing angle. On a phone, turn the brightness up and hold the screen straight in front of you.
How do I retake the same plates?
Press “Same plates again” on the result screen, or use the seed value in the address bar (or the link in the copied result), and the same plates will appear in the same order. This is handy for comparing different lighting or devices.
References
- Regan BC, Reffin JP, Mollon JD (1994). Luminance noise and the rapid determination of discrimination ellipses in colour deficiency. Vision Research 34(10):1279–1299. — Confusion-line directions and the luminance-noise method (Cambridge Colour Test)
- Machado GM, Oliveira MM, Fernandes LAF (2009). A physiologically-based model for simulation of color vision deficiency. IEEE TVCG 15(6):1291–1298. — Simulator matrices
- Sharpe LT, Stockman A, Jägle H, Nathans J (1999). Opsin genes, cone photopigments, color vision, and color blindness. In Color Vision: From Genes to Perception, Cambridge University Press. — Prevalence by type
- Birch J (2012). Worldwide prevalence of red-green color deficiency. JOSA A 29(3):313–320. — Prevalence across populations
- MedlinePlus Genetics (U.S. National Library of Medicine): Color vision deficiency · Achromatopsia · Blue cone monochromacy — Frequency and inheritance of the rarer types