The two first elements can be described in pure physical terms, the
light source by its spectral intensity distribution, the object by its
reflectance spectrum i.e. the fraction of light intensity reflected within
a certain spectral region. The observer imposes limitations since the
electrical signals generated at the eye are processed in the brain through
an unknown process into the final color sensation. Here lies the psychochromatic
origin of color sensation, a point that we always must bear in mind when
dealing with color assessment.
Color science distinguishes also two different steps approaching color
assessment. Both operation are supported in the software:
Color quantification or colorimetric analysis is quite straightforward;
the problem posed by the observer is solved by defining an average observer through experimentally
determined response functions. Color quantification relies on a number
of conventions and definitions given by the CIE
(Commission Internationale de lEclairage) that form the basis for color
quantification and communication. Any color can be characterized by three
numbers, the set of all possible triples define a certain color solid
that contains all possible colors detected by the human eye. Depending
on the system used to quantify color the three numbers have different
names, in general one can say that a color can be characterized by its
saturation, its hue and its luminosity. After the reflectance values of
samples have been measured, they are weighted with those of a light
source and of the observer leading to saturation, hue and luminosity
(or lightness). Two colored samples with exactly the same values for the
color coordinates are said to be equivalent and they are perceived as
equal by the eye under observation
conditions. Bear in mind however that a change in illumination
conditions will lead to different color
coordinates and the color coordinates of color samples may change
in different extent and not be equivalent under the new conditions (they
are said to build a metameric
pair).
There are a lot of color models
or color spaces described in literature. A few of the most commonly used
color models also used in the software have been described in the following:
Color differences or color comparison is a pure industrial problem and
can be stated as to which extent color
coordinates of two samples can differ and still be assessed as the
same by the trained eye; this is the typical color matching problem. There
are many formulas to express color differences but none of them has universal
character i.e. they must be calibrated to the color region where assessment
is conducted.
Color comparison are available under consideration of the following
color models:
