Characterizing Color Properties of LED-Based Reading Lights (2002)

Although the lighting industry routinely uses color rendering index (CRI) to compare the color rendering properties of different light sources, CRI is not a good predictor of object color appearance. Recent LRC research indicated that by slightly shifting the spectrum of the red LED in a red, green, and blue (RGB-Mix) LED system, one can significantly shift the CRI value, from an index of 20 to 70. The LRC hypothesized that although the CRI value is dramatically different for the two RGB-Mix LED light sources, human subject response for color appearance and preference of objects lit by these sources will be almost the same. With funding from the Alliance for Solid-State Illumination Systems and Technologies (ASSIST), the LRC conducted a human factors experiment to verify this hypothesis.

EXPERIMENT

The project evaluated several types of white LED reading lights and compared them to conventional halogen and incandescent reading lights for color rendering and color appearance properties. Human subjects viewed two identical scenes placed side-by-side and lit by different light sources. They rated their preference for a given scene compared to the reference scene. Their evaluations were based on side-by-side comparisons.

CONCLUSIONS

• In general LED-based reading lights were more preferred than halogen and incandescent reading lights.

• Although two RGB-Mix white LED reading lights had significantly different CRI values (23 and 63), subject preference was similar for the two.

• CRI is not a good predictor of object color appearance.

• Phosphor-based white LED reading light was rated poor for the appearance of human skin tones.