Purchase this article with an account.
Caterina Ripamonti, Jakob S. Thomassen; Necessary colour correction for the implementation of the Cambridge Colour Test on a LCD monitor. Invest. Ophthalmol. Vis. Sci. 2016;57(12):191.
Download citation file:
© 2017 Association for Research in Vision and Ophthalmology.
To obtain accurate colour discrimination thresholds with the Cambridge Colour Test (CCT), it is crucial that the display device is able to reproduce colour reliably. To achieve this, the original implementation of the CCT used a dedicated Visual Stimulus Generator (VSG) which is capable of generating chromatic stimuli with 14 bit resolution, on a Cathode Ray Tube (CRT) monitor. As CRTs are no longer made, we tested if a calibrated Liquid Crystal Display (LCD) monitor could offer the equivalent performance, and thus be considered as an alternate stimulus display for the CCT.
We characterised the performance of LCD and CRT displays to reproduce a set of chromatic stimuli, and calculated the just noticeable difference between the measured and requested colour using the delta E value (DE) developed in the CIEDE2000 formula. A value of DE greater than 1 represents a visible difference between the measured and requested colour.
We made a series of systematic measurements which showed that the native performance of LCDs was not as good as the native performance of CRT displays: more than 50% of the colours reproduced by LCDs had a DE greater than 1, compared to only 15% of the colours reproduced by CRT displays. There appear to be several factors that can contribute to this inaccuracy, including (i) variations in the transmittance properties of the LCD due to changes in temperature; (ii) chromatic shift that occurs within the LCD primaries across different intensity levels; (iii) the goodness of the opto-electronic transfer function characterisation and its linearisation; (iv) the contribution of the dark light; and (iv) the colour resolution of the LCD.
We developed an optimised colorimetric correction which takes into account the above factors and found that it can significantly improve the colour reproduction of a calibrated LCD. Using our optimised colour correction algorithm, less than 3% of the colours reproduced by the LCD had a DE>1. This demonstrates that our calibrated LCD has superior colour reproduction compared to the native performance of a CRT display, and provides the confidence that we can re-implement the CCT on this system.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.
This PDF is available to Subscribers Only