Scribd
Upload
16 views24 pages

Representation and Description

Uploaded by

sscreations.official2022155
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
16 views24 pages

Representation and Description

Uploaded by

sscreations.official2022155
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 24

Medical Image Processing II

Representation and Description

References
Digital Image Processing Using MATLAB
RC Gonzalez, RE Woods , SL Eddins

1
Representation and Description

Objects in an image can be represented in terms of


their
1. Boundary characteristics
2. Internal characteristics

2
Boundary Descriptors

Length of a boundary is one of the simplest


descriptor

Length of a 4-connected boundary is equal to the


number of pixels in the boundary minus one

If the boundary is 8-connected we count the vertical


and horizontal transitions as one and diagonal
transitions as √2

3
Boundary Descriptors
Boundary of objects contained in an image is
extracted using the function bwperim
Syntax g = bwperim (f, conn)
Where g is a binary image containing the
boundaries of the objects in f. conn is the
connectivity and it can have the values 4 or 8
(default)
Objects in f can have any pixel values consistent
with image class, but background pixels have to be
0.
Perimeter pixels are nonzero and are connected to
at least one other nonzero pixel
4
Function diameter
Diameter of a boundary is defined as the Euclidian
distance between the two farthest points on the
boundary
These points are not always unique as in a circle or
square
Diameter is a useful descriptor if it applies to a
single pair of points of the boundary
Line segment connecting these points is called the
major axis

5
Function diameter
Minor axis of a boundary is defined as the line
perpendicular to the major axis and of such length
that a box passing through the intersecting points of
the two axes completely encloses the boundary

This box is called the basic rectangle

Ratio of the major to minor axis is called the


eccentricity of the boundary

6
Function diameter

Function diameter computes the diameter, major


axis, minor axis and basic rectangle of a boundary
or region. Its syntax
s = diameter (L)

where L is a label matrix and s is a structure with


the fields Diameter, MajorAxis, MinorAxis and
BasicRectangle

7
Regional Descriptors
Important approach to region description is to
quantify its texture content
There is no existing formal definition for texture
Statistical and spectral approaches can be used
describe texture
Statistical methods give texture as smooth, coarse,
grainy and so on
Spectral methods are based on Fourier spectrum
and used to detect global periodicity in an image

8
Function regionprops
Function regionprops is the principal tool for
computing region descriptors. Its syntax
D = regionprops (L, properties)

where L is a label matrix and D is a structure array

Field of structure denotes different measurements


for each region as specified by properties and they
are shown in next two slides

9
M

10
M

11
Texture Properties

A region can be described by quantifying its texture


content also

Texture content can be computed using statistical


and spectral measures

12
Texture Properties

Texture analysis based on statistical properties of


the intensity histogram is frequently used

One class of such measures is based on statistical


moments

13
Texture Properties
nth moment about the mean intensity is given by

Where zi is a random variable indicating intensity


p(z) is the histogram of the intensity levels in a
region. L is the number of possible intensity levels.
m is the mean (average) intensity

14
Texture Properties

These moments can be calculated using function


statmoments

15
Function statxture
Function statxture computes the texture properties
using the histogram
Syntax t = statxture(f, scale)

Where f is an input image (or sub image) and t is a


6-element row vector whose components are the
descriptors: Mean, Standard deviation, R, 3 rd
moment, Uniformity and Entropy as shown in next
two slides

Function statmoments is used inside the statxture


function
16
Descriptors given by Function statxture

17
Descriptors given by Function statxture

18
Gray Level Co-occurrence Matrix (GLCM)
Measures of texture computed using only histogram
carry no information regarding spatial relationships
between pixels, which is important when describing
texture

It is important to consider not only the distribution of


intensities, but also the relative positions of pixels in
an image

We can achieve this by constructing a GLCM

19
Gray Level Co-occurrence Matrix (GLCM)
Let Q be an operator that defines the position of two
pixels relative to each other and consider an image
f, with L possible intensity levels

Let G be a matrix whose element gij is the number


of times that pixel pairs with intensities z i and zj
occur in image f in the position specified by Q,
where 1 ≤ i, j ≤ L

Matrix formed in this manner is referred as Gray


Level Co-occurrence Matrix
20
How to construct co-occurrence matrix
When L = 8 and Q is defined as “one pixel
immediately to the right”

21
Function graycomatrix
Function graycomatrix computes the gray level co-
occurrence matrixes.
Syntax [GS, FS] = graycomatrix(f, ‘NumLevels’, n,
‘Offset’, offsets)

Most commonly used syntax


G = graycomatrix (f, 'NumLevels', 256);

Usually graycomatrix function is used inside the


graycom function

22
Function graycoprops
Function graycoprops generates the descriptors of
the gray level co-occrence matrix. Syntax

stats = graycoprops(GS, properties)

Where stats is a structure whose fields are the


properties shown in next slide

Usually graycoprops function is used inside the


graycom function

23
Function graycoprops

24

You might also like