Professional Documents
Culture Documents
Lesing 18
Lesing 18
boundary pixels
A segmented region can be represented by
internal pixels
When shape is important, a boundary (external) representation is used
When colour or texture is important, an internal representation is used
The description of a region is based on its representation,
for example a boundary can be described by its length
The features selected as descriptors are usually required to
be as insensitive as possible to variations in (1) scale, (2) translation and (3) rotation, that is the features should be scale,
translation and rotation invariant
11.1 Representation
Image data, for example a boundary, is usually represented in
a more compact way so that it can be described more easily
(d) p4 p6 p8 = 0
N (p1) number of non-zero neighbours of p1
(d) p2 p6 p8 = 0
Delete all the flagged contour points
Repeat steps 1 and step 2 until no contour point is deleted
during an iteration
Reasons for each of these conditions...
0 1 0
(a) N (p1) = 1 : 0 1 0 : end point will be deleted!
0 0 0
1 1 1
N (p1) = 7 : 1 1 1 : erosion will occur!
1 0 1
0 0 1
(b) T (p1) = 2 : 0 1 0 : connectivity will be broken!
1 0 0
Note that N (p1) = 2 here
Reasons for conditions (c), (d), (c) and (d): see page 652
Example 11.1: The skeleton of a region
3
0
2
1
0
2
1
3
2
1
3
0
1 0 1 0 3 3 2 2
3 3 1 3 3 0 3 0
Lecture 18 (p 10)
c
c
When C = 11 12 is the covariance matrix of the boundary
c21 c22
pixel coordinates, the angle that the major axis makes with
the horizontal is given by
y2
c22
= arctan
= arctan 2
c11
x
Construct a basic rectangle in such a way that the boundary
fits within it
length of major axis
Eccentricity of boundary =
length of minor axis
Lecture 18 (p 11)
2 4: eccentricity = 2
For example, options for n = 12: 3 3: eccentricity = 1
1 5: eccentricity = 5
Lecture 18 (p 12)
Lecture 18 (p 13)
(a) Shapes
(b) Hypothetical similarity tree
(c) Similarity matrix