What's new
Photoshop Gurus Forum

Welcome to Photoshop Gurus forum. Register a free account today to become a member! It's completely free. Once signed in, you'll enjoy an ad-free experience and be able to participate on this site by adding your own topics and posts, as well as connect with other members through your own private inbox!

Human Eye Relative Absorption Equliser


bennyh

New Member
Messages
1
Likes
0
Hi, I'm new here,
I'm a neuroscience student learning about the eye with only some hobby experience in Photoshop.
I don't know if anyone has ever seen the graph below (credit to Neuroscience 5th Ed. Purves), but I'd like to make and apply a filter (just out of curiosity) that compensates and for the relative absorption of different wavelengths of the visible spectrum - so that the brightness of the colours across the spectrum are represented equally in the eye. My first thoughts would be to replicate the inverse of this graph in the separate colour channels ... but I'm not sure ...
Any ideas how I might go about this?
Cheers,
Ben
Colour discrimation.PNG
 

thebestcpu

Guru
Messages
2,988
Likes
2,747
Hi Bennyh

I have seen this graph before yet it is not clear what you are truly trying to do and therefore not clear that Photoshop can do what desire.

You are probably way more steeped in this field than I am and many others yet here are some thoughts.

1) The graph you are showing is what I understand is variously called a spectral sensitivity function, luminous efficiency function, relative luminance function etc. The curves you presented are actually just the inverse of curves created when testing the eye know by (as I understand it) "Heterochromatic Brightness Matching" There are mnay references on the internet about this and here is a real simple short one:
http://www.yorku.ca/eye/bright.htm

If all you want is the original curves from which your graph was created, I suggest just going to original source by which is was created.

2) If you wanted to create this curve yourself, it seems a simple matter to put the above curve into a programs such as excel and just do the 1/x function of each point entry of the curve for all four curves and you would also have what you want

3) In the best case, Photoshop could only simulate such curves because in RGB space, Photoshop only works with 3 numbers to simulate a color and cannot work with actual monochrome color frequency wavelengths as is done with Heterochromatic Brightness Matching. Same for virtually all monitors, most are based on just 3 color phosphors at different strengths to make you eye-brain combination see a particular color even though what's presented to you is not monocrhome color by any means. So not sure any Photoshop RGB filter and displaying on an RGB monitor would give you anything of interest academically. In other words, Photoshop could only input 3 unique monocrhome color frequencies albeit each with its own strength.

4) If simulating a monocrhome frequency color via a tri-color input (i.e. RGB), then Photoshop might have a shot of doing something. Each of the above curves present a monochrome output similar to each fo the 3 RGB color channels (they are not colors until interpretted by the brain). So in theory, a monochrome (B&W) filter output could be created to represent the inverse of any individual curve above when stimulated by a tri color input of some overall luminosity. In PS, the maximum luminosity for a given tri-coor is not the same for a given fully saturated color and is determined by L = .3*red + .59*green + .11*blue

--------------------------


So you seemingly simple question is not so simple in Photoshop and pursuing this would really need clarity on what you are trying to achieve.

Not sure this helps yet wanted to share my perspective

John Wheeler



[h=3]
[/h]
 

Top