Abstract
Illumination changes cause alterations in reflectance of object surfaces, while perception of surface colour tends to be constant because of colour constancy. Liquid colour, however, appears drastically different under various chromatic lights. In order to investigate why there is much less effect of colour constancy on liquid colour, we investigated the perception of liquid colour under various chromatic lights. The targets were coloured liquids made by watercolours in a transparent glass container with the dimension of 10 (height) x 10 (width) x 4 (depth) in centimetres. Targets were red, orange, yellow, yellow-green, green, blue, violet, or pink in colour. For each condition, one of the eight coloured targets was placed against a white background ( Munsell value N9 ) and was illuminated by one of six chromatic lights (red, yellow, green, blue, purple, and the illuminant A (ill-A)). The observer was asked to evaluate the colourfulness and hue of the target using an elementary colour naming method, and to evaluate the transparency of the target with the method of magnitude estimation. In order to examine how the illumination colour influences the perception of liquid colour, we compared changes in the colourfulness with the excitation purity in CIE xy-coordinates (1931). The excitation purity of the red target was 20% greater when illuminated by the red light compared to when it was illuminated by the ill-A. On the contrary, the colourfulness was reduced by 70% in the same condition. The reduction is twice as much as the colourfulness for a red paper with the same xy-coordinates as the red liquid target under ill-A. This indicates that there was much less colour constancy effect for the liquid colour than for object surfaces. Differences and similarities with the characteristics of colour constancy between object surfaces and liquid materials may elucidate mechanisms behind the colour appearance for liquid materials.
Meeting abstract presented at VSS 2012