Image Processing Reference
In-Depth Information
(a)
(b)
(c)
(d)
Fig. 10.1
Results of visualization of the moffett
2
data from the AVIRIS using
a
the bilateral
filtering-based technique,
b
the Bayesian technique (reproduced from Fig.
5.3
b
),
c
the variational
technique,
d
the optimization-based technique (reproduced from Fig.
7.2
b
),
e
the three band selection
technique,
f
the piecewise linear function technique, and
g
the color matching function technique.
The gamma is set to 1.50 for all color displays (Color is viewable in e-book only)
the variational fusion technique shown in Fig.
10.1
c is similar to that of the Bayesian
fusion in terms of contrast and clarity in the scene where the smaller objects in the
scene are well identifiable. The fused image using the optimization-based solution
is shown in Fig.
10.1
d. The high amount of visual contrast in this image is due to the
specific objective function in the problem formulation. This image appears somewhat
different in terms of its color composition. However, the coloring scheme being a
pseudo one, it does not reflect the quality of fusion. It may be observed that the band
selection technique essentially captures a very small fraction of data. Therefore, the
corresponding resultant image from Fig.
10.1
e does not contain enough features, and
thus, lacks clarity. This fact will be more evident when this image is compared with
the other results. Figures
10.1
f, g provides the results of fusion of the moffett
2
data
using the piecewise linear functions (PLF), and the color matching functions (CMF),
