# CSCE 5683 Digital Image Processing Midterm Exam Fall 2012

AI and Robotics

Nov 6, 2013 (4 years and 6 months ago)

131 views

CSCE 5683

Digital Image Processing

Midterm Exam

Fall 201
2

Instructions:

This is an in
-
class midter
m

exam.

You are a
llowed one 8.5x11 page of notes.

Raise your hand if you have any questions.

Student Name:_________________________

Student
UAID:_________________________

Question #1

(20 points)

Assume that you are given an input image that is
100
x
100

and you want to create an
output image that is
300
x
300
.

a)

Describe

how
pixel replication

could be used for this process.

b)

Describe how
bilinear

interpolation

could be used for this process.

c)

Question #2

(20 points)

Assume that you are given
a grey scale
medical

image
where
intensity values between
[0..50] are air, [200..255] are

bone, and [50..200] are
tissues of different density.

a)

Use equations and diagrams to d
escribe how
intensity windowing

could be used to
process this
medical
image to
enhance the visibility the tissues in this image.

b)

What would the air pixels and bone
pixels look like in this

windowed image
?

c)

What would the
intensity range

of the image be if you

the original image to
the windowed image above?

Question #3 (20 points)

Low pass filtering is a well known image smoothing technique.

In this question, y
ou will
compare two techniques for low pass filtering.

a)

Describe how
ideal low pass filtering

works.

Draw diagrams to illustrate the
“cross section” of this filter in the frequency domain.

b)

Describe how
Butterworth low pass filtering

works.

Draw diagrams to
illustrate
the “cross section” of this filter in the frequency domain.

c)

Give one advantage of the ideal low pass filter over the Butterworth filter.

d)

Give one advantage of the Butterworth low pass filter over the ideal filter.

Question #
4

(20 points)

Consid
er the
8x8

image below.

a)

Calculate the
intensity histogram

for this image h(i)
.

b)

Calculate

the
cumulative intensity histogram

for this image
H(i).

c)

Based on this information, do you think the pixels with intensity 2 would be
larger or smaller if we perform

histogram equalization on this image?

0

0

0

0

1

2

4

8

0

0

0

0

1

2

4

8

0

0

0

0

1

2

4

8

0

0

0

0

1

2

4

8

1

1

1

1

1

2

4

8

2

2

2

2

2

2

4

8

4

4

4

4

4

4

4

8

8

8

8

8

8

8

8

8

Question #
5

(20 points)

Consider the 8x8 image below.

It has the same pixel
values as the previous image but

to a small number of pixels
.

a)

Describe how you could
remove this noise

using
median filtering

using a 3x3
neighborhood.

Illustrate the algorithm with one
3x3 neighborhood

below
.

b)

Describe ho
w you could remove this noise using
binomial filtering

using a 3x3
neighborhood.

Illustrate the algorithm with one 3x3 neighborhood below.

0

0

0

0

1

2

4

8

0

0

5

0

1

2

4

8

0

0

0

0

1

2

4

8

6

0

0

0

1

8

4

8

1

1

1

1

1

2

4

8

2

2

2

2

2

2

4

1

4

4

0

4

4

4

4

8

8

8

8

8

8

8

8

8