Geoscience Reference
In-Depth Information
h e size of
image1
is 666-by-329-by-3, while
image2
is slightly smaller: 614-by-
270-by-3. We display the images by typing
subplot(1,2,1), imshow(image1)
subplot(1,2,2), imshow(image2)
h e function
imregconfig
creates optimizer and metric coni gurations that
we transfer into
imregister
to perform intensity-based image registration,
[optimizer, metric] = imregconfig('monomodal');
where
monomodal
assumes that the images were captured by the same sensor.
We can use this coni gurations to calculate the spatial transformation matrix
tform
using the transformation type
affine
, as in the previous example.
tform = imregtform(image2(:,:,1),image1(:,:,1), ...
'affine',optimizer,metric);
h is transformation can be applied to
image2
in order to automatically align
it with
image1
.
a
b
c
Fig. 8.4
Automatically aligned TERRA-ASTER images of the Suguta Valley in the Northern
Kenya Rit ;
a
i rst image taken on 20th February 2003,
b
second image taken on 31st August
2003, and
c
overlay of the second image aligned with the i rst image. Original image courtesy
of NASA/GSFC/METI/ERSDAC/JAROS and U.S./Japan ASTER Science Team.