jsColorblindSimulator

Usage

Use the file selector in the top left, drag and drop or paste an image here to load it. You will then be able to choose between different simulations for different kinds of color-blindness.

All processing is done locally in your web browser. The image will not leave your computer.

The displayed image can be moved by clicking and dragging and zoomed by (shift-)double clicking or using the mouse wheel.

The lens-feature allows you to see what the original image looked like in the region around your cursor. The inverse lens shows you the original image and the simulated image around your cursor.

Uses algorithms from

• http://www.colorjack.com/labs/colormatrix/ (web.archive link, because original source is gone)
• http://www.nofunc.com/Color_Blindness_Library (web.archive link, because original source is gone)
• libDaltonLens (implementation of Computerized simulation of color appearance for dichromats by Brettel et al. 1997)
• Implementation of A Physiologically-based Model for Simulation of Color Vision Deficiency by Machado et al.

The first two algorithms have not been proven to be accurate.

Comment about the first algorithm:

You're right, the ColorMatrix version is very simplified, and not accurate. I created that color matrix one night (http://www.colorjack.com/labs/colormatrix/) and since then it's shown up many places... I should probably take that page down before it spreads more! Anyways, it gives you an idea of what it might look like, but for the real thing...

As far as a simple script to simulate color blindness, this one does the best job:

http://www.nofunc.com/Color_Blindness_Library/ - It uses "confusion lines" within the XYZ color space to calculate values (this one is in Javascript, and should be easy to convert to python).

There are a few other methods, and no one really knows exactly what it would look like... these are all generalizations of a small sample, set against the masses.

Copyright Notice for the second non-commercial algorithm:

The Color Blind Simulation function is
copyright (c) 2000-2001 by Matthew Wickline and the
Human-Computer Interaction Resource Network ( http://hcirn.com/ ).

It is used with the permission of Matthew Wickline and HCIRN,
and is freely available for non-commercial use. For commercial use, please
contact the Human-Computer Interaction Resource Network ( http://hcirn.com/ ).

Comment about the third algorithm:

The Brettel et al. method was published in 1997 in the age of CRT monitors and has been adapted to modern sRGB monitors. It should be pretty accurate, at least for full dichromacy. Of course it is still an approximation and many factors make it imperfect, including an uncalibrated monitor, unknown lighting environment, and per-individual variations. In general it will tend to be accurate for small or thin objects (small dots, lines) and too strong for large bright areas, even for full dichromats.