Abstract
This paper analyzes spectral reflectance information in order to understand color features. Twenty-three-color image data are taken by using a monochrome CCD (Charge Coupled Device) camera through three filters: X, Y and Z filters on condition that the range of illumination is 10001x to 110001x at every 10001x, and that a lens opening is eight types. Digital number (luminous intensity) in image data is much related to both brilliance and illumination at data acquisition. The relation between spectral reflectance information and brilliance will help extract the difference of colors. Based on the measurements and the resulting relations, we propose a color distinction approach due to fuzzy reasoning. The reliability of the proposed approach is demonstrated in an experiment using the color image data that taken in a XYZ space.
