Purchase this article with an account.
Joao Linhares, Sergio Nascimento; Rendering Ishihara color plates on a computer screen using hyperspectral images: will the scores be the same as the traditional paper test?. Journal of Vision 2013;13(9):1024. doi: 10.1167/13.9.1024.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
The use of hyperspectral imaging data to render digital images enables accurate lighting simulations and color accuracies unattainable with more traditional methods of image acquisition and digital rendering. The goal of this work was to use hyperspectral data to test how accurate is the Ishihara test displayed on a calibrated computer monitor.
Hyperspectral images of twenty-three plates of the Ishihara test were acquired from 400-720nm in 10nm steps with high spatial resolution. The spectral data of the paper that encircled the colored plates and the lighting spectrum of the light box were also independently measured. These data were used to generate the monitor simulations of the Ishihara plates.
32 color normal subjects did the test using the Ishihara book and the corresponding images on a calibrated monitor screen. The observers’ task was the traditional one of reading the numbers printed on the Plates. Their color vision was also assessed using the Heidelberg Anomaloscope.
Only one test was performed on each session being the book test the first one and the screen test the second one. In the screen test the plates were presented in randomized order.
25% of the observers made no mistakes on the paper or screen test. 37.5% made one mistake performing better on the screen test. 12.5% made 2 errors performing better on the paper test. 25% had more than 3 errors performing better on the paper test.
These results suggest that the use of hyperspectral images and the spectral information of the illumination to render Ishihara color vision can be used to produce an equivalent Ishihara test on a calibrated computer screen.
Meeting abstract presented at VSS 2013
This PDF is available to Subscribers Only