Scribd
Upload
55 views4 pages

MATLAB Based Image Processing Lab Experiments: Session S3F

This document describes a set of MATLAB-based image processing experiments for a digital image processing course. It includes 13 proposed experiments that cover topics like point transformations, morphological operations, histogram equalization, Fourier transforms, filtering, and compression. Examples are provided for thresholding and histogram equalization experiments. The median filter experiment shows how a median filter can effectively remove impulse noise from an image. The experiments are designed to help students learn and apply fundamental image processing concepts.

Uploaded by

Armando Garcia Mejia
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
55 views4 pages

MATLAB Based Image Processing Lab Experiments: Session S3F

This document describes a set of MATLAB-based image processing experiments for a digital image processing course. It includes 13 proposed experiments that cover topics like point transformations, morphological operations, histogram equalization, Fourier transforms, filtering, and compression. Examples are provided for thresholding and histogram equalization experiments. The median filter experiment shows how a median filter can effectively remove impulse noise from an image. The experiments are designed to help students learn and apply fundamental image processing concepts.

Uploaded by

Armando Garcia Mejia
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

Session S3F

MATLAB Based Image Processing Lab Experiments


Carlos Sandoval Zuria, Juan Manuel Ramirez, David Bciez-Ldpez,
and G. Espinosa Flores- Verdad'
Departamento de Ingenieria Electrdnica
Universidad de las Ame'ricas-Puebla
Cholula, Puebla 72820
'Departamento de Electrdnica
Instituto Nacional de AstroJisica, Optica y Electrdnica
Tonantzintla,Puebla
MEXICO
[email protected]
Abstract - This paper presents a set of experiments for
a digital image processing course. This set of
experiments is based on MATLAB. These experiments
will allow any instructor to cover experiments in most
of the topics treated in a regular image processing
course to be completed successfully.

I. Introduction
Motivation for a digital image processing course [ 1-31
arises mainly from two areas of applications, namely, a)
an increase in the pictorial information available for
human interpretation, and b) image processing for
automatic and autonomous machine control. This is
because vision is the most important human sense in
terms of the amount of information it conveys and
because of a good visualization is very important for the
correct information converged in an image.
Many image processing circuits require dedicated
software to perform their tasks. These packages usually
are highly priced and are not easily modified by the final
user. In this paper the basis for the image processing is
MATLAB [4], a software package now available almost
anywhere for other uses and that is used as the engine for
the image processing experiments. The software written
for the image processing experiments is available from
the users at no cost to interested users and instructors.
Image processing. applications include many topics,
among which we can mention remote sensing,
ultrasound images, meteorology, astronomy, inspection,
radar, seismology, radiology, autonomous navigation,
recognition, etc.

images can be thought of as matrices. A definition for an


image can then be the following:
An image is a NxN array of elements. Each element in
the array is a number which represents the sampled
intensity. The samples are named pixels (picture
element). After sampling each pixel is quantized. Each
intensity is assigned a number within a finite set of
values, usually between 0 and K- 1, where K=2B is the
possible number of gray levels, each represented by B
bits. The digitizing of images is now complete and each
image is now an array which can be handled more
appropriately as a matrix.
Our Image processing laboratory based on MATLAB
has the capability to perform the following tasks:
1.

Perform algebraic operations such as addition,


subtraction, multiplication, and division. These
operations can be used to perform operations on
images such as noise reduction by using averages,
movement detection, and algebraic masking.

2.

Geometric transformations such as translation,


rotation, and scaling

3.

Space domain operations such as histogram


modification (scaling, offset, amplitude change) non
linear point operations (absolute value, squaring,
square root, log scale compression, edge detection)

4.

Binary image processing (thresholding,


operations).

5.

Non linear image processing such as morphologic


operators (opening, closing). Structuring element
choice. Dilation, erosion.

6.

Frequency domain processing. Fourier transform,


log compression.

11. Image Processing Laboratory


MATLAB is a matrix oriented computing engine.
Thus, it is almost perfect for image processing because
++o*p,,

0-7803-4762-5/98/$10.00 91.998 IEEE

:
~4

1998 FIE Conference

=%*,plj
.,H..*
1255

98CH36214

logic

7.

Filtering by linear convolution. Filter design (low


pass, high pass, band pass, band reject.). Gaussian
Filters. Linear restoration. White noise non linear
filtering.

8. Digital Image coding


and
Compression measures, losses
entropy, optimal coding.

compression.
compression,

111. Suggested List of Experiments

10. Ideal filters in the frequency domain. This


experiment allows students to appreciate the effects
of filtering low and high frequencies in an image.

11. Non Linear filtering using convolutional masks.


This experiment allows students to understand the
effects of a median filter on an image corrupted with
impulsive noise.

This section provides a list of experiments which cover


the topics of the previous section.

12. Entropy as a compression measure. This


experiments introduces students to entropy as a
compression measurement to the DPCM
compression measure.

Point-to-point transformation. This laboratory


experiment provides for thresholding an image and
the evaluation of its histogram. The user can choose
a threshold level to see the image showing only the
pixels at that threshold.

13. Edge detection. This experiment enables students


to understand the concept of edge detectors and their
operation in noisy images.

1.

2.

Morphological operations I. This experiment is


intended so students can appreciate the effect of
morphological operations using a small structuring
element on simple binary images. The operations
that can be performed are erosion, dilation, opening,
closing, open-close, close-open.

3 . Morphological operations 11. This experiment is


designed to let students know how morphological
functions change images by applying consecutive
erosion and dilation operations.

4.

Histogram
equalization.
This
experiment
illustrates the relationship among the intensities
(gray levels) of an image and its histogram. It
shows how to improve the image by equalizing the
histogram.

5.

Geometric transformations. This experiment shows


iinage rotation, scaling, and translation.

6.

Two-dimensional Fourier transform I. The purpose


of this experiment is to provide an understanding of
the harmonic content of an image using the discrete
Fourier transform (DFT).

7 . Two-dimensional Fourier Transform 11. This


experiment is designed so the student learns the
concept o f masking with the DFT.
8. Linear filtering using convolution. After completing
this experiment every student should understand the
concepts of filtering using linear convolution.
9.

Highly selective filters. In this experiment students


appreciate the effects on an image after a highly
selective filter is applied to it.

0-7803-4762-5/98/$10.00 Q 1998 BEE

It is worth mentioning that in some of these


experiments, the functions defined have been completely
written in MATLAB code and in other cases use have
been made of internal MATLAB functions such as rand,
show-img, ffQ, ifft2, fftshift, etc.

IV. Examples
Our first example is related to experiment No. 1 about
image enhancement by use of point-to-point
transformations. The goal of enhancement techniques is
to process a given image so that the result is more
suitable than the original image. An histogram provides
statistical information about the use of the range of gray
levels in a digital image (e.g., the histogram of a dark
image is concentrated in he loer gray levels). Fig. 1 a
shows a typical Mexican church tower and Fig. l b
shows its histogram. Here students can observe that
most of the pixel's intensities are between 120 and 200.
In other words, this image has a small dynamic range.
To obtain further information about the image we can
choose a value of the gray level and observe the image
formed by pixels at that value. For example, choosing
first as a threshold the value of 129 and then the value of
140, the resulting images in Figs. 2a and 2b are
obtained. Note that clearly there are more pixels with
intensity 140 and, therefore, Fig. 2b has more
definition. Note that Fig. 2b shows a glimpse of a tower
while Fig. 2a does not. Further processing can be done
with the histogram. This is done in the next example.
A related experiment is Experiment No. 4 for histogram
equalization. The purpose of histogram equalization is
to optimize the use of the dynamic range. Fig. 3 shows
an original image and its histogram. Note that both
ends of the histogram are not used meaning that there
are neither black nor white pixels in the figure. Fig. 4
shows the result on the image after histogram
equalization. It can be noticed the contrast in the
I

1998 FIE Conference

1256

98CH36214

equalized image now that the range of gray levels is


completely covered.

(a)

Histogram ofthe image

I '

(b)
50

150

100

200

Intensity

250

Fig. 2. a) Threshold at 129.


b) Threshold at 140.

b)
Fig. 1. a) Mexican church tower.
(b) Histogram for Fig. 1a.
Fig. 5 shows the church tower from Fig. 1 a k r
applying histogram equalization. Note that the contrast
improves drastically since in the original image grey
levels are clustered around the center values and now
they are distributed throughout all the range. This
example shows how a simple concept can be used to
improve images. This is in general a concept for non
linear signal processing.

Our third example (experiment No. 11) shows the effects


of a median filter on an impulsive noise corrupted
image. The function of a median filter is to replace the
value of each pixel by the median of the gray levels in a
neighborhood of that pixel. The use of median filters to
remove impulsive filters is particularly effective. The
size of the neighborhood to evaluate the median is
defined by the user. Fig. 6 shows the noisy image and
the restored image. (peppers).

p*p*,,

0-7803-4762-5/98/$10.006>1998 IEEE

:agj;
I,
48..

*. s

1998 FIE Conference

1257

98CH36214

Histogram of the sdbimage

2130,

Fig. 5. Equalized church tower. The image contrast has


improved after equalization.

Fig. 3 Portion of Lena and its histogram.

Histogram of the equalized image

200,

Ib)
Fig., 6 . Noise removal by applying non linear filtering.
a) 11mage with impulsive noise added, b) filtered image .
Fig. 4 This figure shows image from Fig. 3 after
histogram equalization.

References

Conclusions
A set of image processing experiments based on
MATLAB has been developed. This set of experiments
covers most of the topics in a regular image processing
course. An additional use of this set would be to use it
in a regular Digital Signal Processing course, where
image processing occupies a small part, to show digital
image processing applications.

[ l ] A.C. Bovik, Digital Image Processing Course


Notes, Dept. of Electrical Engineering, U. of Texas at
Austin, 1995.
[2] J.C. Russ, The Image Processing Handbook, CRC
Press, Boca Raton, FL., 1992.

[3] R.C. Gonzalez and R.E. Woods, Digital Image


Processing, Addison-Wesley, Reading, Mass., 1992.

[4] MATLAB, v. 4.0, The Mathworks, Inc., Natick,


MA., 1995.

1998 FIE Conference

0-7803-4762-5/98/$10.00 0 1998 IEEE


I"0

1258

I
E
E

98CH36214

You might also like