Professional Documents
Culture Documents
Chapter 11 (Feature Extraction)
Chapter 11 (Feature Extraction)
Chapter 11 (Feature Extraction)
and Description
Image Representation and Description?
Objective:
To represent and describe information embedded in
an image in other forms that are more suitable than the
image itself.
Benefits:
- Easier to understand
- Require fewer memory, faster to be processed
- More “ready to be used”
What kind of information we can use?
- Boundary, shape
- Region
- Texture
- Relation between regions
Representation & Description
Segmentation techniques yield raw data in the
form of pixels along a boundary or pixels
contained in a region.
After segmentation, the image needs to be
described and interpreted.
Representation: an object may be represented by its
boundary.
Description: the object boundary may be described by
its length, orientation, or number of concavities...
Representation
An object can be represented by:
Its
external characteristics, such as its
boundary.
X X
X X
X X
Description
The features that represent the image are used
as descriptors.
representation boundary
Step 4: Mark the current pixel as previous pixel & the next pixel as
the current pixel.
Step 5: Go to step 3
Boundary (Border) Following …
S
P 1C 1 1 1
1 1
1 1
1 1
1 1 1 1
Boundary (Border) Following …
S
1P 1C 1 1
1 1
1 1
1 1
1 1 1 1
Boundary (Border) Following …
S
1 1P 1C 1
1 1
1 1
1 1
1 1 1 1
Boundary (Border) Following …
S
1 1 1P 1C
1 1
1 1
1 1
1 1 1 1
Boundary (Border) Following …
1 1 1 1P
1 1C
1 1 S
1 1
1 1 1 1
Boundary (Border) Following …
1 1 1 1
1 1P
1 1C
1 1 S
1 1 1 1
Chain Codes
Represent a boundary of a connected region
Why focus on a boundary?
The boundary is a good representation of an object shape
and also requires less memory.
Chain codes: it represents an object boundary by a
connected sequence of straight line segments of specified
length and direction.
The direction of each segment is coded by using a
numbering scheme shown below
1101101030332330322212
Chain Codes …
22120207656764443
Chain Codes …
Unacceptable because
the resulting chain of codes tends to be quite long
33001122
33001122 00112233
30011223 01122330
00112233 11223300 First row gives the
01122330 Sort 12233001 normalized chain code
11223300 rows 22330011 00112233
12233001 23300112
22330011 33001122
23300112 30011223
The First Difference of a Chain Codes
Problem of a chain code:
a chain code sequence depends on a starting point.
Solution: treat a chain code as a circular sequence and redefine the
starting point so that the resulting sequence of numbers forms an
integer of minimum magnitude.
33001212 33010122
01011311
01011311 01131101
Shape Numbers
Order :8
Shape Number …
1
Shape number of the boundary definition:
the first difference of smallest magnitude
2 0
The order n of the shape number:
the number of digits in the sequence
3
Shape Number …
Shape Number …
1. Original boundary
Chain code:
000030032232221211
First difference:
300031033013003130
4. Find the nearest
Shape No.
3. Create grid Grid.
000310330130031303
Polygon Approximation
Represent an object boundary by a polygon
Minimum perimeter
Object boundary
polygon
Minimum perimeter polygon consists of line segments that
minimize distances between boundary pixels.
Polygon Approximation:Splitting Techniques
1. Find the line joining
two extreme points
0. Object boundary
2. Find the
farthest points
from the line
3. Draw a polygon
Distance-Versus-Angle Signatures
2D boundary ⇒ 1D function
Distance from centroid as function of angle
not applicable to all shapes
Distance-Versus-Angle Signatures
Represent an 2-D object boundary in term of a 1-D function
of radial distance with respect to q.
Distance-Versus-Angle Signatures
1 K 1
a (u ) s(k )e 2uk / K
K k 0
Reconstruction formula
1 K 1
Boundary s(k ) a (u )e2uk / K
points K k 0
Example: Fourier Descriptor
Examples of reconstruction from Fourier descriptors
1 P 1
sˆ( k ) a (u )e 2uk / K
K k 0
P is the number of
Fourier coefficients
used to reconstruct
the boundary
Eccentricity
Ratio of the major to the minor axis
Major axis = the line connecting the two extreme
points that comprise the diameter
Minor axis = the line perpendicular to the major
axis
Statistical Moments
Definition: the nth moment
K 1
n ( r ) ( ri m)n g ( ri )
i 0
Example of moment:
where K 1 The first moment = mean
m ri g ( ri ) The second moment = variance
i 0
Boundary
segment
1D graph
Area
Perimeter
Compactness
Topological descriptors
Texture
Regional Descriptors …
Regional Descriptors …
- Area of the region: the number of pixels in the region
- perimeter = length of its boundary
- Compactness
A( R) Example: a circle is the most
C 2 compact shape with C = 1/4
P ( R)
where A(R) and P(R) = area and perimeter of region R
Topological Descriptors
Use to describe holes and connected components of the region
E = -1
E=0
Any Questions ?