Abstract
To investigate the tolerances in the CIE System of assessing colour rendering differences, seventy-two observers have compared two sets of the eight Munsell samples recommended in the System. Each set was illuminated by one of a pair of fluorescent lamps of different colour rendering properties selected from six of similar correlated colour temperature (4000 K). It is shown that the CIE method is not applicable to the comparison of lamps unless one lamp is the reference standard. The observations were recorded on a 4-category scale and were transformed by means of a theoretical model to a subjective ratio scale which is linearly related to the measured colour difference corrected for adaptation of the eye. Furthermore, the just noticeable colour differences were approximately five-and-a-half times the 'minimum perceptible colour differences' of MacAdam's observer. A 'just noticeable' special colour rendering index could thereby be derived for individual samples.
Get full access to this article
View all access options for this article.