Color Blindness Test

How can I check for colour blindness online with an Ishihara-style test?

Open AnyTool Color Blindness Test, set your screen to full brightness with any night-light or colour filter off, then run the plates: each is a circle of speckled dots with a number hidden in a colour that sits on a colour-vision confusion line. Type the digits you can read, or tap “I can’t see a number”, and move through the set. At the end it gives a cautious screening verdict — such as “responses consistent with a possible red-green deficiency” or “none detected” — and shows which axis the misses cluster on. It is a screening and awareness tool only, not a diagnosis: screen colours vary so much that it cannot confirm the type or severity. Every plate is generated procedurally on a canvas, with no copyrighted images, and the whole tool runs in your browser.

Generated plates — no copyrighted Ishihara images, drawn on canvas
Type the number you see, or say you can’t see one
Probes both red-green (protan/deutan) and blue-yellow (tritan) axes
Cautious screening verdict + which axis the misses cluster on
Screening only — NOT a diagnosis; 100% in your browser, nothing uploaded

What is

Colour vision deficiency screening (Ishihara-style test)

A colour vision deficiency (CVD) screening uses pseudoisochromatic plates — circles of coloured dots with a figure hidden in dots whose colour lies on a confusion line — to flag whether someone may struggle to tell certain colours apart. A person with a matching deficiency sees little contrast between the figure and the background and reads the number wrong or not at all. The most common form is red-green deficiency (protan and deutan, ~8% of men and ~0.5% of women); blue-yellow (tritan) deficiency is rare. On-screen plates can screen for a possible deficiency but cannot diagnose the type or severity, which needs a printed Ishihara booklet under standard lighting or an anomaloscope.

Calculators

Related terms

Colour blindnessIshihara testPseudoisochromatic plateProtanopiaDeuteranopiaTritanopiaConfusion lineAnomaloscope

Frequently Asked Questions

Online plates can only ever suggest a possible colour vision deficiency. Screen colours, brightness and calibration vary enormously between devices, and a display has just three primaries that it reproduces inconsistently, so it cannot reproduce the precisely measured printed inks a clinical test relies on. That means an online test cannot tell protan from deutan, cannot grade severity, and can miss milder cases. Use it for awareness, then see an optometrist or ophthalmologist for a proper Ishihara booklet under standard lighting or an anomaloscope before deciding anything that matters.

No. It is a screening and awareness tool that can flag a possible deficiency, but it cannot diagnose the type or severity.

Each plate is a pseudoisochromatic plate: a circle filled with randomly sized dots, where the dots forming the figure are coloured from one family and the background dots from a confusable family. When the two families lie on a colour-vision confusion line — for example a warm orange figure against an olive-green ground for red-green probes — a person with that deficiency perceives them as nearly the same colour and cannot pick out the number, while a colour-typical viewer reads it easily. AnyTool generates these plates procedurally on a canvas rather than using copyrighted plate art.

A number is hidden in dots whose colour sits on a colour-vision confusion line, so people with that deficiency see little contrast and misread it.

Deutan (weak or missing green M-cone) is the most common type, affecting about 6% of men; protan (weak or missing red L-cone) affects about 2% of men, and like deutan it confuses reds, greens, browns and oranges, but also dims reds. Together these red-green types affect roughly 1 in 12 men and 1 in 200 women, and are X-linked, which is why they are far rarer in women. Tritan (weak or missing blue S-cone) is a rare blue-yellow deficiency that affects both sexes equally because it is not sex-linked. Complete colour blindness (monochromacy) is very rare.

Red-green (protan and deutan) is by far the most common; blue-yellow (tritan) is rare. Total colour blindness is very rare.

The whole test depends on subtle colour contrast, and a screen reproduces colour far less consistently than printed ink. Low brightness, a night-light or blue-light filter, a colour-blind assist mode, or coloured ambient lighting can all shift the apparent colours enough to change the outcome. For the least-misleading run, set your display to full or near-full brightness, disable any colour filters or night mode, and sit in neutral indirect light. Even then, treat the result as a rough screen, not a measurement — the colours can never match a printed Ishihara booklet exactly.

Because the plates are sRGB approximations — every screen renders the colours differently, so wrong brightness or a colour filter can change your result.

An Ishihara-style colour vision screening that generates pseudoisochromatic dot plates procedurally on a canvas — probing both the red-green (protan/deutan) and blue-yellow (tritan) confusion axes — and returns a cautious screening verdict. A screening and awareness tool that runs entirely in your browser, not a medical diagnosis.

Features

Procedurally generated plates on canvas — no copyrighted Ishihara images
Type the number you see, or report that you cannot see one
Probes red-green (protan/deutan) and blue-yellow (tritan) confusion lines
Cautious screening verdict with which axis the misses cluster on
Honest screening-not-diagnosis limits — 100% client-side, nothing uploaded

PriceFree

CategoryCalculator

PlatformAll (browser-based)

PrivacyNothing leaves your device — plates are generated and scored in your browser

Input: Typed digit answers per plate, Optional “I can’t see a number”

Output: Screening verdict (possible red-green / blue-yellow / none detected), Per-plate breakdown, Per-axis misses

Detailed Explanation

Methodology

How Procedurally Generated Ishihara-Style Plates Work

AnyTool Color Blindness Test builds every plate in the browser — no copyrighted Ishihara art is used or downloaded. A deterministic engine fills a disc with hundreds of randomly placed, varied-size dots using a seeded PRNG, so each plate renders identically every time. The figure is a digit rendered through a compact 5×7 bitmap font into a mask; any dot whose centre lands on the mask is coloured from a “figure” palette, and the rest from a “background” palette. The two palettes are chosen to sit on a colour-vision confusion line — for red-green probes a warm orange/brown figure against an olive-green ground, for blue-yellow probes a teal figure against a pink/grey ground — so a person with that deficiency perceives little contrast and misreads the number, exactly the pseudoisochromatic principle of the printed test. The dots are painted onto a device-pixel-ratio-aware canvas clipped to the disc.

Plates generated on canvas — no copyrighted images, nothing downloaded
Seeded PRNG → identical, stable plates every render
Digit drawn via a 5×7 bitmap font into a figure mask
Figure vs background palettes sit on a colour-vision confusion line
Red-green (warm vs olive) and blue-yellow (teal vs rose) probes

How It Works

Types of Colour Vision Deficiency and Their Prevalence

Colour vision deficiency comes in three main families. Deutan (a weak or missing green M-cone) is the most common, affecting about 6% of men; protan (a weak or missing red L-cone) affects about 2% of men and, like deutan, confuses reds, greens, browns and oranges while also dimming reds. Together these red-green deficiencies affect roughly 1 in 12 men and 1 in 200 women and are X-linked, which is why they are far more common in men. Tritan (a weak or missing blue S-cone) is a rare blue-yellow deficiency that affects both sexes about equally because it is autosomal, not sex-linked. The screening set therefore weights red-green probes heavily, includes a couple of blue-yellow probes, and opens with a demo plate that nearly everyone can read so a missed demo flags a screen problem rather than a deficiency.

Deutan (green) most common — ~6% of men
Protan (red) ~2% of men; both confuse red-green and dim reds
Red-green deficiency ~1 in 12 men, ~1 in 200 women; X-linked
Tritan (blue-yellow) rare; affects men and women equally
A demo plate detects screen/lighting problems vs true deficiency

Limitations

Why This Is Screening, Not a Diagnosis

The result is a cautious screening verdict, never a diagnosis. A clinical Ishihara test relies on inks measured under standard daylight illumination; a screen has only three primaries it reproduces inconsistently from device to device, and brightness, night-light filters, colour-blind assist modes and ambient lighting all shift the apparent colours. So the tool cannot tell protan from deutan, cannot grade severity, and can miss milder cases. The scoring clusters misses by axis and reports outcomes such as “responses consistent with a possible red-green deficiency”, “possible blue-yellow issue”, “inconclusive — check your screen” (when the demo plate is missed), or “none detected” — and a none-detected result does not rule out a mild deficiency. The page states this prominently up front, repeats it next to the result, and directs users to an optometrist for a printed Ishihara booklet or an anomaloscope.

Screen colour varies; plates are sRGB approximations, not printed inks
Cannot distinguish protan vs deutan or grade severity
Missed demo plate → inconclusive, likely a screen/lighting problem
“None detected” does not rule out a mild deficiency
Confirm with an Ishihara booklet or anomaloscope via a clinician

Privacy & Security

Privacy and Compatibility

The Color Blindness Test is 100% client-side: plates are generated and painted, answers are read, and the verdict is computed entirely in the browser with no server, account or tracking, and it works offline after first load. Because plates are drawn on a canvas rather than fetched as images, nothing about your answers or your colour vision ever leaves your device. It runs on phones, tablets, laptops and monitors, though a colour-accurate display at full brightness gives the least-misleading run.

Online colour blindness screening: AnyTool vs typical web colour tests
Capability	AnyTool	Typical online colour tests
Processing	Runs in your browser	Runs in your browser
Plate source	Generated on canvas, no copyrighted art	Often reused copyrighted plate images
Axes probed	Red-green and blue-yellow, labelled	Often unlabelled or red-green only
Demo / sanity plate	Yes — flags screen problems	Rarely
Answer method	Type the number or say none	Often multiple-choice (guessable)
Result detail	Verdict + per-axis + per-plate	Pass/fail only, if any
Calibration honesty	Brightness reminder + sRGB caveat up front	Often omitted
"Not a diagnosis" framing	Stated prominently and repeated	Often understated
Cost / signup	Free, no signup	Often ad-heavy or gated

AnyTool generates every plate locally and uploads nothing — but it is a screening aid, not a medical diagnosis.

Cite This Page

Title: Color Blindness Test (Ishihara-style colour vision screening with procedurally generated pseudoisochromatic plates, red-green and blue-yellow probes, no server)
Author: AnyTool
Published: January 1, 2026
Modified: February 8, 2026
Section: Calculators
URL: https://www.anytool.tech/color-blindness-test

Formatted Citation

AnyTool. "Color Blindness Test (Ishihara-style colour vision screening with procedurally generated pseudoisochromatic plates, red-green and blue-yellow probes, no server)." AnyTool, January 1, 2026. https://www.anytool.tech/color-blindness-test

A privacy-first, 100% client-side colour vision deficiency screening tool that runs entirely in the browser and is explicitly a screening and awareness aid — NOT a medical diagnosis. Every plate is generated procedurally on a canvas with no copyrighted Ishihara images: a seeded PRNG fills a disc with hundreds of varied-size dots, a digit is rendered through a 5×7 bitmap font into a figure mask, and dots on the mask take a figure palette while the rest take a background palette, with the two palettes chosen to sit on a colour-vision confusion line (a warm orange/brown figure against olive-green ground for red-green protan/deutan probes; a teal figure against pink/grey ground for blue-yellow tritan probes) so a person with that deficiency sees little contrast and misreads the number. The user types the number they see or reports they cannot see one; the set opens with a demo plate almost everyone can read, weights red-green probes (the most common deficiency — deutan ~6% of men, protan ~2%, together roughly 1 in 12 men and 1 in 200 women, X-linked) and includes blue-yellow probes (tritan, rare, affects both sexes equally). Scoring clusters misses by axis into a cautious verdict — possible red-green, possible blue-yellow, possible mixed, inconclusive (demo missed → likely a screen problem), or none detected — and the page states prominently and repeatedly that screen colour, brightness and calibration vary so much that online plates are unreliable: the tool cannot distinguish protan from deutan, cannot grade severity, can miss mild cases, and a none-detected result does not rule out a deficiency. Users are urged to confirm anything that matters with an optometrist using a printed Ishihara booklet under standard lighting or an anomaloscope. A brightness/colour-filter reminder is shown before testing, and nothing is uploaded.