Results and Discussions - Hyperspectral Image Fusion - page 164

Fig. 10.6 Results of visu-
alization of the urban data
from the Hyperion using a
the bilateral filtering-based
technique, b the Bayesian
technique (reproduced from
Fig. 5.2 b ), c the variational
technique, d the optimization-
based technique (reproduced
from Fig. 7.1 b ), e the three
band selection technique, f
the piecewise linear function
technique, and g the color
matching function technique

(a) (b)

(c) (d)

This procedure was completed in 320 s for the AVIRIS datasets, and in 98 s for the
Hyperion datasets on the same computer system described earlier. The second part
of the technique is estimation of the fused image from the so obtained -factors.

This being an iterative procedure, the exact time for the convergence of the solution
varies with the data, and the parameters of the algorithm. The average time required

β