A study of the variation in color matching functions with respect to field size and primary set.

摘要

The primary purpose of the study was to derive the CMFs for a large field of view, and to determine whether there are systematic differences among CMFs as field size varies. In addition, Thornton found experimentally that CMFs do not transform appropriately between primary sets. Therefore, the second purpose of this study was to explore whether the primary set has significant impact upon CMFs.; The UNL colorimeter was designed and color matching experiments were performed by ten subjects. Each subject completed experiments for three different matching conditions: a 10° field of view with the primary set PC (named 10PC), a large field of view with PC (named LFPC), and a 10° field with the primary set NP (named 10NP). PC and NP are two sets of primary lights with the peak wavelengths of 453-533-601 and 478-553-636 nm respectively. The large field of view has 102° (wide) x 50° (high) dimensions.; The results show that the field size has significant impact upon the ratio of luminance in the reference field to that in the test field and on the resulting CMFs. The LFPC has lower sensitivities than 10PC in the short wavelengths of the b¯ curve, and has higher sensitivities in the long wavelengths of the r¯ curve.; The study also shows that the primary set has a significant affect on the ratio of power in the reference field to that in the test field, the color difference between the reference field and test field, and the resulting CMFs. The 10NP CMFs have lower sensitivities than the 10PC CMFs in the g¯ curve between 460 and 540 nm.; Our experiments confirmed Thornton's conclusion regarding the transformation between primary sets. We found that transformation with different transformation matrices or with different primary systems can result in different results.