Purchase this article with an account.
Derek Nankivil; On the Assessment of the Chromatic Contrast of Ecologically Valid Stimuli. Invest. Ophthalmol. Vis. Sci. 2018;59(9):4041. doi: https://doi.org/.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Model the chromatic contrast of ecologically valid stimuli and evaluate the role of spectral filtering.
For any object, we define a vector in the perceptually uniform CIE L*a*b* color space. With this, the relative perceptual differences between any two objects was estimated as the Euclidean distance between the two colors in L*a*b* space. Then, following the Michelson contrast, we obtain an expression for chromatic contrast in CIE L*a*b*: Blue-yellow (BY) and red-green (RG) contrasts were calculated in an ecologically valid scenario using published reflectance or irradiance spectra for zenith skylight, stage 4 Cavendish bananas, Summerred apples, apple tree leaves, and direct sunlight at standard temperature and pressure. Spectra were converted to XYZ tristimulus values using the CIE standard observer functions, and then to L*a*b* using the D65 white point as a reference. Perfect high pass filters with cutoffs ranging from 390 to 550 nm were simulated in 1 nm increments and the filtered chromatic contrast was calculated for both the BY and RG stimuli.
The unfiltered BY and RG stimuli presented chromatic contrasts of 0.872 and 0.870, respectively. A high pass filter with a cutoff at 410 nm provided a maximum in BY contrast of 0.873, improving BY contrast by less than 0.1%, while the RG contrast remained unchanged at 0.870. A high pass filter with a cutoff at 435 nm provided a maximum in RG contrast of 0.878, improving RG contrast by less than 1%, while decreasing BY contrast by more than 8%.
A model of chromatic contrast of ecologically valid stimuli was developed and applied to the assessment of high pass spectral filters. Differences in the response of each stimuli to spectral filtering suggest that the ideal cutoff filters may vary from scene to scene.
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.
This PDF is available to Subscribers Only