3 ms·

Ask HN: Are you colour blind? Please help validate daltonize algorithm in Ruby

8 points by gingerlime 14y ago

Looking for colour-blind people to help validate the implementation of the daltonize algorithm in ruby (against the python / javascript implementations). The results look slightly different but are they just as effective / better / worse? Any assistance will be appreciated. see some samples and discussion on https://github.com/eltiare/carrierwave-vips/issues/9#issuecomment-13685362

logn 14y ago
I believe I'm a Strong Deutan. I'm reviewing your deuteranope set at that link you provided. The Python colors look a lot different than the original but I see the numbers the same in those two, with the exception of the middle right image, the "5" and the middle bottom, the "15". Each of those in the original is less defined and the 5's look closer to 8's for me (although I can still tell it's a 5). With the Ruby, every one of your numbers is hazier and harder to read with the exception of the top-left, the "12". However, I believe if given this test, my answers on the Ruby version would only deviate from the original for the middle-right (5), bottom-left (3), and middle-bottom (15). Also it seems the tan and yellow hues are more accentuated in the Ruby version than the original, but the colors are a lot closer than the Python version. I can give my review of the other tests if you want but with my color-blindness, I don't believe my opinion is relevant for those.
- gingerlime 14y ago
  Thanks logn. Much appreciated. Just to clarify a couple of points if you don't mind: "(although I can still tell it's a 5)" - that is - on the original? or only the python version? If you are able to see (almost) all numbers on the original, then perhaps your colour-blindness is not as strong? or the test isn't great to start with?? "However, I believe if given this test, my answers on the Ruby version would only deviate from the original...". Deviate meaning you won't be able to tell the difference? Would it be possible to contact you via email/skype to follow-up on this if necessary?
  - logn 14y ago
    >"(although I can still tell it's a 5)" - that is - on the original? or only the python version? That's regarding the original. The numbers I see in the original (left->right, top->bottom) are: 12, 8, 6, 29, 57, 5, 3, 15, 74. Is that what you see? They're designed so colorblind people see different numbers sometimes. If that's exactly what you see then the test is either flawed or it's not inclusive of all the frames which test this colorblindness. I have severe color-blindness and take these at the eye-doc regularly. I also read the original color-blindness book in biology class once which is how I know my type. >"However, I believe if given this test, my answers on the Ruby version would only deviate from the original...". Deviate meaning you won't be able to tell the difference? Deviate in that on the Ruby version my answers might be: 12, 8, 6, 29, 57, _3_, _8_, _17_, _71_ Feel free to email me: pleasedofeelfreetosendmespam@gmail.com ... yes that's my email :)
  - logn 14y ago
    Also, for reference, all the samples here: http://www.toledo-bend.com/colorblind/Ishihara.asp http://www.toledo-bend.com/colorblind/Ishihara.asp Except for the top-left 25, they're all spots for me. The 56 in the middle-right is maybe perceptible to me, but barely.
    - gingerlime 14y ago
      Thanks again. That's cool. I'll try to pass those tests through the algorithm (both python and ruby) and share the results somewhere - hopefully it will be a better test case then. > If that's exactly what you see then the test is either flawed or it's not inclusive of all the frames which test this colorblindness Yes - that's what I see, so maybe these samples weren't the best to start with. I took them straight off http://daltonize.appspot.com/ http://daltonize.appspot.com/ so assumed they are a good starting point...
gingerlime 14y ago
Another version (thanks to logn) - available at http://imgur.com/a/nCzap http://imgur.com/a/nCzap
- logn 14y ago
  Based on this result I think the Ruby algorithm is performing pretty well. Python makes the first four fairly readable for me while Ruby is truer to the original. However the Python and Ruby hues vs original are still a good bit different (but I'm not exactly sure if that's what you want or not)
devonbarrett 14y ago
I just had my brother test them out. I am not too sure what he suffers with but they defiantly work - I would say even more effectively than the original! The one he found hardest to see was the python one.

8 points by gingerlime 14y ago

logn 14y ago

I believe I'm a Strong Deutan. I'm reviewing your deuteranope set at that link you provided. The Python colors look a lot different than the original but I see the numbers the same in those two, with the exception of the middle right image, the "5" and the middle bottom, the "15". Each of those in the original is less defined and the 5's look closer to 8's for me (although I can still tell it's a 5). With the Ruby, every one of your numbers is hazier and harder to read with the exception of the top-left, the "12". However, I believe if given this test, my answers on the Ruby version would only deviate from the original for the middle-right (5), bottom-left (3), and middle-bottom (15). Also it seems the tan and yellow hues are more accentuated in the Ruby version than the original, but the colors are a lot closer than the Python version. I can give my review of the other tests if you want but with my color-blindness, I don't believe my opinion is relevant for those.

gingerlime 14y ago
Thanks logn. Much appreciated. Just to clarify a couple of points if you don't mind: "(although I can still tell it's a 5)" - that is - on the original? or only the python version? If you are able to see (almost) all numbers on the original, then perhaps your colour-blindness is not as strong? or the test isn't great to start with?? "However, I believe if given this test, my answers on the Ruby version would only deviate from the original...". Deviate meaning you won't be able to tell the difference? Would it be possible to contact you via email/skype to follow-up on this if necessary?
- logn 14y ago
  >"(although I can still tell it's a 5)" - that is - on the original? or only the python version? That's regarding the original. The numbers I see in the original (left->right, top->bottom) are: 12, 8, 6, 29, 57, 5, 3, 15, 74. Is that what you see? They're designed so colorblind people see different numbers sometimes. If that's exactly what you see then the test is either flawed or it's not inclusive of all the frames which test this colorblindness. I have severe color-blindness and take these at the eye-doc regularly. I also read the original color-blindness book in biology class once which is how I know my type. >"However, I believe if given this test, my answers on the Ruby version would only deviate from the original...". Deviate meaning you won't be able to tell the difference? Deviate in that on the Ruby version my answers might be: 12, 8, 6, 29, 57, _3_, _8_, _17_, _71_ Feel free to email me: pleasedofeelfreetosendmespam@gmail.com ... yes that's my email :)
- logn 14y ago
  Also, for reference, all the samples here: http://www.toledo-bend.com/colorblind/Ishihara.asp http://www.toledo-bend.com/colorblind/Ishihara.asp Except for the top-left 25, they're all spots for me. The 56 in the middle-right is maybe perceptible to me, but barely.
  - gingerlime 14y ago
    Thanks again. That's cool. I'll try to pass those tests through the algorithm (both python and ruby) and share the results somewhere - hopefully it will be a better test case then. > If that's exactly what you see then the test is either flawed or it's not inclusive of all the frames which test this colorblindness Yes - that's what I see, so maybe these samples weren't the best to start with. I took them straight off http://daltonize.appspot.com/ http://daltonize.appspot.com/ so assumed they are a good starting point...

gingerlime 14y ago

Another version (thanks to logn) - available at http://imgur.com/a/nCzap http://imgur.com/a/nCzap

logn 14y ago
Based on this result I think the Ruby algorithm is performing pretty well. Python makes the first four fairly readable for me while Ruby is truer to the original. However the Python and Ruby hues vs original are still a good bit different (but I'm not exactly sure if that's what you want or not)

devonbarrett 14y ago

I just had my brother test them out. I am not too sure what he suffers with but they defiantly work - I would say even more effectively than the original! The one he found hardest to see was the python one.