Spectral Analysis of Omnidirectional Illumination in a Natural Scene

Abstract

This article describes a method for analyzing omnidirectional color signals in a natural scene, which contains direct illuminations of daylights and indirect illuminations of the reflected lights from different object surfaces. A multiband omnidirectional imaging system is used for capturing high resolution images in the omnidirectional observations at a particular point in a natural scene. The spectral-power distributions of color signals are recovered from the captured six-band images. The authors investigate the spectral composition of the omnidirectional scene illumination based on the principal component analysis of the whole set of color signals acquired at the same location in a fixed time of day in five months of a year. It is found that all the omnidirectional color signals can be expressed in a linear combination of only three principal components. This property has the potential for high data compression. Moreover, the authors analyze the chromaticity distribution of omnidirectional color signals. Experimental results are presented for omnidirectional color signals obtained in an outdoor scene of university campus. The reliability of the proposed method is examined from various points of view. An application to image rendering is shown.