Pages: [1]   Go Down

Author Topic: Any simple way to demonstrate "flicker photometry"?  (Read 867 times)

rasworth

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 473
Any simple way to demonstrate "flicker photometry"?
« on: June 01, 2018, 05:46:55 pm »

I would like to teach an elementary colorimetry class at our local photography club meeting.  Toward this end I want to demonstrate the "flicker photometry" experiment using a pc/projector.  The technique is one used to measure the eye's relative sensitivity to different colors, by alternating two different narrow spectrum colors projected into the same area at a frequency of approximately 17 hz.  The colors will blend to create a third color, but if the perceived "brightness" of each color is different then a flicker will be apparent.  One color is projected at a fixed intensity, and the observer varies the amplitude of the other color until no flicker appears.  In theory one can "hop" across the visible spectrum and derive the human eye photopic luminosity function.

I'm not looking for anything accurate, just trying to get the concept across.  Anybody have an idea how to alternate two layers in Photoshop using a script, or whatever, to demonstrate?  I searched for an app, there are plenty with RGB sliders to synthesize different colors, but nothing I found could be adapted for "flickering".

Richard Southworth
Logged

Doug Gray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2197
Re: Any simple way to demonstrate "flicker photometry"?
« Reply #1 on: June 01, 2018, 05:55:37 pm »

You should be able to demonstrate it using animated gif images. You can create them in Photoshop. To create them you should use two colors, each with an XYZ value that added together form white. You can find opposing colors in a standard RGB space that do so using Bruce Lindbloom's color calculator.
Logged

rasworth

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 473
Re: Any simple way to demonstrate "flicker photometry"?
« Reply #2 on: June 01, 2018, 07:50:40 pm »

Doug, thanks, I'll give it a try.

Richard Southworth
Logged

rasworth

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 473
Re: Any simple way to demonstrate "flicker photometry"?
« Reply #3 on: June 02, 2018, 05:53:07 pm »

Close, but I can't get the frequency up enough to demonstrate the effect, tops out at 10 hz or so.  There's also the issue of syncing with the display refresh.  I tip my hat to what those 1920 scientists accomplished with the technology at hand!

Richard Southworth

« Last Edit: June 03, 2018, 10:46:49 am by rasworth »
Logged

rasworth

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 473
Re: Any simple way to demonstrate "flicker photometry"?
« Reply #4 on: June 03, 2018, 02:13:27 pm »

This almost works, using very small images and Firefox for viewing.  Color1 and Color2 are equal L*, 85, in sRGB.  Color3 has L* = 89.  TestEQ.gif uses Color1 and Color2, TestDIFF.gif Color1 and Color3.  Unfortunately I wasn't able to push the frequency high enough so that TestEQ.gif was a completely non-flickering color, but there is still a significant difference when compared to TestDIFF.gif.

Richard Southworth

Added by edit - I tried 4 browsers, Firefox and Chrome seemed to work, IE and Edge show both gifs as blinking with no discernible difference.
« Last Edit: June 03, 2018, 02:22:06 pm by rasworth »
Logged

rasworth

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 473
Re: Any simple way to demonstrate "flicker photometry"?
« Reply #5 on: June 03, 2018, 02:14:22 pm »

Here's Color3.

Richard Southworth
Logged
Pages: [1]   Go Up