A Perceptual Study of Linear Models of Spectral Reflectance

It is known that the reflectance spectra of both natural and man-made surfaces may be represented efficiently using linear models. A key task is to answer how many basis functions of a linear model are necessary for a given accuracy of representation. While the most common application is the reproduction of color and images, which highly relate to human perception, it is important to understand the relationship between human perceptual tolerance of spectral images and the number of basis functions of the corresponding linear models. In this study several data sets and psychophysical studies have been used to address the question of how many basis functions are necessary to represent a spectral image so that it is indistinguishable from the original (complete) representation. The results show that the number of basis function required is highly image dependent: 3 to 8 basis functions needed for different types of images.