
296 Variational Methods in Image Processing
FIGURE 12.8: Deformed template (left), deformation field plus the land-
marks transformation (right) and 1/det(∇Φ) plus the deformed grid (bottom)
using biharmonic regularization.
sity function of X and Y, and p
X
(x) and p
Y
(y) are the marginal probability
density functions of X and Y, r e spe c tively.
Mutual information quantifies the dependence between X and Y, which
in our case are the intensity maps of R and T . Considering that larger mu-
tual information indicates better regis tration, we see in Figure 12.10 that the
nonlinear e lasticity model indicated by the black line that is above is more
desirable