Page 300 -

techniques commonly employed in the handling and manipulation of digi-
tal images.
Let us explore and observe the structure of a matrix subjected to the above
elementary trasformations. For this purpose, execute and observe the out-
puts from each of the following commands:

M=(1/25)*[1 2 3 4 5;6 7 8 9 10;11 12 13 14 15;16
17 18 19 20;21 22 23 24 25]
lrM=fliplr(M)
udM=flipud(M)
Mr90=rot90(M)

A careful examination of the resulting matrix elements will indicate the gen-
eral features of each of these transformations. You can also see in a visually
more suggestive form how each of the transformations changed the image of
the original matrix, if we render the image of M and its transform in false col-
ors, that is, we assign a color to each number.
To perform this task, choose the colormap(hot) command to obtain the
images. In this mapping, the program assigns a color to each pixel, varying
from black-red-yellow-white, depending on the magnitude of the intensity at
the corresponding detector.
Enter, in the following sequence, each of the following commands and at
each step note the color distributions of the image:

colormap(hot)
imagesc(M,[0 1])
imagesc(lrM,[0 1])
imagesc(udM,[0 1])
imagesc(Mr90,[0 1])
The command imagesc produces an intensity image of a data matrix that
spans a given range of values.

9.3.2 Digital Image Processing
A typical problem in digital image processing involves the analysis of the
raw data of an image that was subject, during acquisition, to a blur due to the
movement of the camera or to other sources of noise. An example of this sit-
uation occurs in the analysis of aerial images; the images are blurred due,
inter alia, to the motion of the plane while the camera shutter is open. The
question is, can we do anything to obtain a crisper image from the raw data
if we know the speed and altitude of the plane when it took the photograph?
The answer is affirmative. We consider for our example the photograph of
a rectangular board. Construct this image by entering:

© 2001 by CRC Press LLC