SEPARATING A COLOR SIGNAL INTO ILLUMINATION AND SURFACE REFLECTANCE COMPONENTS - THEORY AND APPLICATIONS

被引:68
作者
HO, JA [1 ]
FUNT, BV [1 ]
DREW, MS [1 ]
机构
[1] BELL NO RES,OTTAWA K1Y 4H7,ONTARIO,CANADA
关键词
Color constancy; color perception; computer vision; finite-dimensional linear model; statistical methods;
D O I
10.1109/34.58869
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
We present a “separation” algorithm for achieving color constancy and theorems concerning its accuracy. The algorithm requires extra information from the optical system, over and above the usual three values mapping human cone responses. However, with this additional information—specifically, a sampling across the visible range of the reflected, color-signal spectrum impinging on the optical sensor—we are able to separate the illumination spectrum from the surface reflectance spectrum contained in the color-signal spectrum which is, of course, the product of these two spectra. At the heart of the separation algorithm is a general statistical method for finding the “best” illumination and reflectance spectra, within a space represented by finite-dimensional linear models of statistically typical spectra, whose product closely corresponds to the spectrum of the actual color signal. Using this method, we are able to increase the dimensionality of the finite-dimensional linear model for surfaces to a realistic value. One method of generating the spectral samples required for the separation algorithm is to use the chromatic aberration effects of a lens, and an example of this is given. We detail the accuracy achieved in a large range of tests and show that agreement with actual surface reflectance is excellent. © 1990 IEEE
引用
收藏
页码:966 / 977
页数:12
相关论文
共 32 条
[21]   COLOR CONSTANCY - A METHOD FOR RECOVERING SURFACE SPECTRAL REFLECTANCE [J].
MALONEY, LT ;
WANDELL, BA .
JOURNAL OF THE OPTICAL SOCIETY OF AMERICA A-OPTICS IMAGE SCIENCE AND VISION, 1986, 3 (01) :29-33
[22]  
MALONEY LT, 1985, THESIS STANFORD U ST
[23]   AN UPDATED VERSION OF THE KANTOROVICH-THEOREM FOR NEWTON METHOD [J].
MIEL, G .
COMPUTING, 1981, 27 (03) :237-244
[24]  
MORE JJ, 1980, ANL8074 TECH REP
[25]   CHARACTERISTIC SPECTRA OF MUNSELL COLORS [J].
PARKKINEN, JPS ;
HALLIKAINEN, J ;
JAASKELAINEN, T .
JOURNAL OF THE OPTICAL SOCIETY OF AMERICA A-OPTICS IMAGE SCIENCE AND VISION, 1989, 6 (02) :318-322
[26]  
Sallstrom P., 1973, 7309 U STOCKH I PHYS
[27]   COUNTING METAMERIC OBJECT-COLOR STIMULI USING FREQUENCY-LIMITED SPECTRAL REFLECTANCE FUNCTIONS [J].
STILES, WS ;
WYSZECKI, G ;
OHTA, N .
JOURNAL OF THE OPTICAL SOCIETY OF AMERICA, 1977, 67 (06) :779-784
[28]   THE SYNTHESIS AND ANALYSIS OF COLOR IMAGES [J].
WANDELL, BA .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1987, 9 (01) :2-13
[29]  
WANDELL BA, 1986, COLOR RES APPL, V11, pS30
[30]   NECESSARY AND SUFFICIENT CONDITIONS FOR VONKRIES CHROMATIC ADAPTATION TO GIVE COLOR CONSTANCY [J].
WEST, G ;
BRILL, MH .
JOURNAL OF MATHEMATICAL BIOLOGY, 1982, 15 (02) :249-258