Image Processing Reference
In-Depth Information
difficult. Generally, unsupervised method is used to cluster the images that
automatically find the structures in the data set.
7.8.1 Colour Model
Colour models are the mathematical representation of a set of colours. There
are different types of colour models such as RGB (red, green, blue), HSV (hue,
saturation, value) and CIELab. In CIELab, the three coordinates are
L
*,
a
*,
b
*:
L
* = 0 indicates black and
L
* = 100 indicates white,
a
*- negative values indi-
cate green, positive values indicate magenta, and
b
*- negative values indicate
blue, positive values indicate yellow. RGB is the common representation of
colour image due to its simple representation. It is device dependent and is
not a perceptual model. CIELab colour space is a perceptually human model
where the Euclidean distance between two colour points is equivalent to the
difference between two colours as perceived by humans. It is denoted by
three Cartesian coordinates:
L
*,
a
* and
b
*. To obtain a CIELab colour model
from RGB, two steps are required: First, it transforms the RGB to CIE
XYZ
colour model and then CIE
XYZ
to CIELab using the following formulas [12].
Transformation of RGB to CIE
XYZ
XR
=
0490310
.
+
.
G
+
0 200
.
B
Y
=
0 177
.
R
+
0 813
.
G
+
0 010
.
B
Z
=
0 000
.
R
+
0 010
.
G
+
0 990
.
B
Transformation of CIE
XYZ
to CIELab
(/)
13
⎛
⎜
Y
Y
⎞
⎟
⎛
⎜
Y
Y
⎞
⎟
>
116
−
16
for
0 008856
.
*
N
N
L
=
⎛
⎜
Y
Y
⎞
⎟
⎛
⎜
Y
Y
N
⎞
⎟
≤
0 008856
903 3
.
for
.
N
⎡
⎢
⎤
⎥
⎛
⎜
X
X
⎞
⎟
−
⎛
⎜
Y
Y
⎞
⎟
a
*
=
500
f
f
N
N
⎡
⎢
⎤
⎥
⎛
⎜
Y
Y
⎞
⎟
−
⎛
⎜
Z
Z
⎞
⎟
b
*
=
200
f
f
N
N
where
(/)
13
t
,
t
>
0 008856
.
ft
()
=
16
116
7 787
.
t
+
,
otherwise
Search WWH ::
Custom Search