Color vision reference plates
Read a single digit inside a freshly packed circle of dots. These original reference exercises use three fixed model confusion directions. After answering, compare the original plate with three transforms and review responses by direction.
Key features
- Original dot plates built from Canvas digit masks and spatial packing
- Three fixed model directions with original and transformed comparisons
- Ten questions with digit answers, unsure choices and correctness
- Response counts by direction, separating unassisted answers
- A contrast hint that reveals the mask and digit
- A separate 3×3 hue and HSL lightness tile exercise
How to use
- Choose number plates or tiles. Changing exercise or settings clears the current session.
- For plates, choose mixed directions or one model direction, then start ten questions.
- Enter the single digit you see, or choose unsure or a contrast hint.
- After submitting, compare the answer and the original plate with three transforms.
- Continue through the questions and review the reference pattern table and your session observations.
Use cases
- Read digits in freshly generated dot arrangements
- Compare how identical color pairs change across models
- Review your responses by reference direction
- Use hints to understand the digit mask
- Practice hue and lightness differences separately with tiles
Frequently asked questions
Is this a clinical test or an Ishihara plate?
No. The reference exercises are authored from Canvas text masks and packed circles. They do not copy Ishihara plates or answer sequences. They cannot determine color vision deficiency, normal vision or driving and service eligibility.
What do the three directions and table values mean?
The tool uses fixed severity-100 Machado et al. 2009 transforms. Color pairs were selected to converge under each transform and checked with CIE76 color distance. This is a numerical plate check, not a clinical validation or personal diagnosis from confusion lines. Residual differences remain, particularly under the tritan transform.
Do digits and dot layouts change?
New sessions use secure randomness for digits and layout seeds. Dot positions and sizes are independent of the digit mask; only their colors change. The fixed example starts with 3 and reproduces its layout. Text rendering varies slightly across devices.
How are unsure choices and hints recorded?
Unsure is counted separately and never treated as correct. A hint increases contrast and reveals the digit in text, so that answer is marked assisted. You can review correct, unassisted correct and unsure counts by direction as well as each response.
Can displays change the results?
Yes. Brightness, color management, night mode, ambient light, viewing angle and screen size all matter. The plate surround stays neutral in both themes, but the display is not calibrated. Transforms do not guarantee a reproduction of any individual’s view.
What happens to tile practice and saved data?
The separate tile mode retains 3×3 hue and HSL lightness comparisons at three authored differences. Neither mode has a time limit. Switching modes clears the session. Inputs and results remain only in page memory, without upload or automatic storage.
Privacy
Settings, selected tiles and results stay in this page’s memory. Nothing is automatically saved or sent to a server or tool analytics events. Starting again, changing settings or clearing the run discards its history.
Comments & questions