Image Processing Reference

In-Depth Information

updating of weights is continued until the network stabilizes. The method is said
stable
(the

learning stops) when:

where
E
(
t
) is the method fuzziness computed at the
t
th iteration,
γ
is a prefixed very small

positive quantity and . After convergence, the pixels
j
with
o
j
>
θ
are con-

sidered to constitute the skin map of the image; they are set to take value 255, in contrast with

the remaining which will constitute the background (value 0).
Figure 5
shows, with a 3D rep-

resentation, the segmentation process (b) and (c) performed by the MS-RNN method on the

input image depicted in (a).

FIGURE 5
(a) Original images, (b)
L
component, (c)
a
component, and (d) skin map (SM).

4 Skin Map Segmentation

We transformed the image model into CEI
Lab
and normalized the luminance component
L

and the chrominance components
a
in the range of [0, 255]. To realize multiclass image seg-

mentation, the
CRC
must satisfy a homogeneity constraint, i.e., the difference between
c
i
,
j
1
and

c
i
,
j
2
must be less than or equal to a prefixed threshold. In such a case, the region is seen as uni-

form and becomes
RC
otherwise, the
CRC
is split into four newly defined
CRC
, letting
w
be

w
/2. The parameters of the preprocessing subnetwork have been set to the following values:

•
w
0
= 8,
w
t
=
w
t
− 1/2
(
t
denotes iteration)

•
θ
= 50

The output of the preprocessing subnetwork normalized in the [0, 1] range is fed to a clus-

tering subnetwork. The parameters of the clustering subnetwork have been set to the follow-

ing values:

•
η
= 0.2 (learning rate)

•
γ
= 0.001 (convergence rate)

The reason for these choices resides in a most successful skin-detection system, both for fed

tecting skin and suppressing noise, while requiring the minimum amount of computation or,

equivalently, minimum number of iterations to converge.

Search WWH ::

Custom Search