Image Processing Reference
In-Depth Information
2
3
q
n
2
=
n
1
2
2
2
i
ð
n
2
=
n
1
Þ
cos
i
ð
Þ
sin
4
5
p
?
¼
q
n
2
=
n
1
(
10
:
55
b)
2
2
2
i
ð
n
2
=
n
1
Þ
cos
i
þ
ð
Þ
sin
p
k
þ
p
?
2
p
1
n
1
, n
2
, i
ð
Þ ¼
(
10
:
55
c)
where the parallel and perpendicular subscripts refer to the polarization of the
incident beam. Hence, the total re
ectance, p
1
(n
1
, n
2
, i), can be simply calculated
by averaging p
k
and p
?
, as described by Equation 10.55c. It is known that the FSR
dilutes the colored light re
ected from within the object, thus lowering the perceived
chroma [32]. According to the same model, gloss affects the direction rather than the
magnitude of the re
ected light. The perceived color depends on how much of the
surface re
ected light is detected. Some instruments, for example, integrating
spheres, detect virtually all of this light, in which case the measured color is
independent of the sample gloss. On the contrary, instruments with 0
=
=
45 or 45
0
(or generally, 0
ected light,
making the perceived color a strong function of gloss; the glossier samples have a
larger specular component and, consequently, the perceived chroma is higher. The
captured portion of the re
=
D or D
=
0) geometry, reject almost all specularly re
ected light (which is a fraction of the FSR p
1
(n
1
, n
2
, i)) is
de
ned as p
0
(g). It is expected to be a decreasing function of gloss, which by
fitting a
logistic curve to data was found to be
1
p
0
(
g
) ¼
22
p
1
(
n
1
, n
2
, i
)
(
10
:
56
)
2
:
1
þ (
g
=
18
:
55
)
It is important to point out that in order to calculate the gloss using the fusing
model described above, the total TMA has to be used; that is, we need to add the
TMAs for the different toner layers:
TMA ¼ TMA
C
þ TMA
M
þ TMA
Y
þ TMA
K
(
10
:
57
)
This is based on the standard assumption that the different toners have similar
thermal properties, hence when the fusing model is used to determine the surface
gloss, only the total mass is important and not the individual masses.
The part of the light that is not re
ected, that is, [1
p
1
(n
1
, n
2
, i)], is transmitted
through the toner layers, re
ected by the substrate (paper), and then transmitted
back through the toner (Figure 10.29). This gives rise to the factor [1
p
1
(n
1
, n
2
, i)]
t
0
(m
i
;
t
0
(m
i
;
t
(m
i
;
l
)R
s
(
l
)
l
) in Equation 10.54. For generality, the term
l
) is not
considered to be equal to
). Indeed, they are not always equal since in practice
transmittance depends on the order of the different layers through which light passes.
For example, transmittance is different when light passes through the yellow layer
followed by the magenta layer than vice versa. Even if the transmittance is con-
sidered to be approximately the same regardless of layer order for the case described
above, this will not be the case for halftoned images (see Section 10.3). This is
t
(m
i
;
l
Search WWH ::
Custom Search