Main menu
User login
Syndicate
You are here
Color Vision Deficiency
Simulates different types of color vision deficiency (color blindness) with differing levels of severity.
In addition to the functionality of the display filter "Color Vision Deficiency", this plug-in also simulates color weakness, not just complete color blindness and allows to save the resulting image.
- Protanomaly (red weak)
- Deuteranomaly (green weak)
- Tritanomaly (blue weak)
At a severity level of 1.0, it simulates dichromacy, i.e. the complete absence of the relevant color receptor:
- Protanopia (red blind)
- Deuteranopia (green blind)
- Tritanopia (blue blind)
The plug-in takes an RGB* image, and simulates for the active layer how a color-deficient viewer would experience it. The result is saved in a new layer.
Currently, the plug-in appears under the top-level menu "Python-fu/Effects/Color Vision Deficiency".
Based on http://www.inf.ufrgs.br/~oliveira/pubs_files/CVD_Simulation/CVD_Simulati...
| Attachment | Size |
|---|---|
 color_vision_deficiency.py.txt | 7.74 KB |
- Log in to post comments
Comments
channel mixer ?
It's slow. How about a channel mixer approch ?
re: channel mixer
not sure it can be implemented with a channel mixer, since it's actually a non-trivial transformation of the color space (different colors become similar and some similar colors become different, for deuternopia red and green get mapped to the same 'color').
it was fast enough for my purposes, some inline-c could prob. speed it up manifold at the cost of deployability. i can look into it for faster processing, but think that it's doing ok for applying a color space transformation matrix on every pixel in python.
i'd be very happy about any suggestions regarding performance.
How about a filter to
How about a filter to simulate red/green contrast blindness to simulate animal (dog, deer, cow, etc) vision?
L*a*b decompose, fill "*a" channel to 50% grey, recompose.
The L and b channels can be manipulated to closer reflect what the animals perceive.
animal color perception
i like the idea, but don't know very much about animal color perception. if you could provide me with some more details and maybe even some scientific sources I would love to add this functionality.
Google the subject,
Google the subject, especially note animal hunting camo colour vision.
I'm not a big fan of hunting, but they have put a lot of effort into figuring out how animals see the world.
Some frequency filtering (high pass of a scale and then set layer to grain extract from the original) may be necessary to show detectable objects.
neat
This will be interesting to try out.
It's already there for ages
Er... Actually GIMP has a built-in display filter exactly for that.
display filter
As far as I could see, the display filter only simulates full color blindness, not color weakness. (6.3% of the population have anomalous trichromacy (color weakness), while only 2.4% have full dichromacy (color blindness)).
Also, I don't know how to save a picture with the display filter applied, short of taking a screenshot.
Not really. You can switch
Not really. You can switch between protanopy etc.
re: not really
Yes, that's what I was saying. As far as I could see you can "only" choose between protanopia, deuteranopia and tritanopia in the display filter. Those three vision impairments are cases of dichromacy - i.e. a 2-dimensional color space instead of a 3-dimensional one. 2.4% of the population suffer from dichromacy.
This plug-in also allows to simulate anomalous trichromacy which is a much weaker from of the above vision impairments. 6.3% of the population suffer from anomalouss trichromacy, which still represents an albeit slightly transformed 3-dimensional color space.