PROJECT TITLE PROJECT TITLE PROJECT

TITLE PROJECT TITLE PROJECT TITLE

This report is submitted to

Silicon Institute of Technology, Bhubaneswar
for
Practice School

Submitted by
Name of the Student (0801209086)

Under the Esteemed Supervision of

Prof. Name of Guide

DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING

SILICON INSTITUTE OF TECHNOLOGY
SILICON HILLS, BHUBANESWAR – 751024, ODISHA, INDIA
Month, Year
 CERTIFICATE

As given by the concerned Practice School.

Date :

Place:

Prof. Aaaaaa Kuuuu Daaaaa

Designation

DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING

SILICON INSTITUTE OF TECHNOLOGY
BHUBANESWAR – 751024
 DECLARATION

We hereby certify that:-

a. The work contained in the report is original and has been done by me under the
supervision of our supervisor.

b. The work has not been submitted to any other Institute for any degree or diploma.

c. We have conformed to the norms and guidelines given to us by the Project

Review Committee of our department.

d. Whenever we have used materials (data, theoretical analysis and text) from other
sources, we have given due credit to them by citing them in the text of the
project and giving their details in the references.

Date :

Place:

Name of the Student (0801209086) …………(signature)…...

DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING

SILICON INSTITUTE OF TECHNOLOGY
BHUBANESWAR – 751024
 ACKNOWLEDGEMENTS

Write an acknowledgement for maximum of one page. The candidates should convey
their appreciation to all whom have played a role for completion of this project work.
The supervisor, co-supervisor (if any), head of the department, faculty members, lab
mates, friends etc. may be acknowledged.

For example, I take upon this opportunity to acknowledge the many people
whose …………… I am deeply indebted to my supervisor and mentor
…………………. I would like to thank ……………… I further thank to
………………….

I owe my sincere gratitude towards …………………. My heartfelt thanks to

……………….. I also express my deepest gratitude to …………………. Finally, I
would like to wind up by paying my heartfelt thanks ………………

Name of the Student …………………………………...

DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING

SILICON INSTITUTE OF TECHNOLOGY
BHUBANESWAR – 751024
 ABSTRACT

Abstract should be of around 250 words. It should be written in Times New Roman, font
size 12, text justified, line spacing 1.5. Writing an effective abstract is very important and
plays an important role because it demonstrates the general idea of the whole work.
Abstract should highlight the project. The objectives and findings should be correlated
and clear. The brief summary of the project with details of highlighting various chapters
should be included in the abstract. The introduction and literature review are not required
to incorporate in the abstract part of the project.

Abstract should be compacted within one or two pages of around 250 words. At
the end of the abstract, eight to ten keywords should be mentioned which are separated
by commas. Keywords are the important buzz words of the project.

Keywords: January, February, March, April, May, June, July, August, September.
 LIST OF ABBREVIATIONS

(Abbreviations should be alphabetically written)

Abbreviation Description

AIR All India Radio

BARC Bhabha Atomic Research Centre

CNTs Carbon Nanotubes

DMSO Dimethyl Sulfoxide

ETA Estimated Time of Arrival

HRTEM High Resolution Transmission Electron Microscopy

IoT Internet of Things

M. Tech Masters of Technology

OS Operating System

PVA Polyvinyl Alcohol

RMSE Root Mean Square Error

SEM Scanning Electron Microscopy

SIT Silicon Institute of Technology

XPS X-ray Photoelectron Spectroscopy

 LIST OF FIGURES

Page #
Chapter 1.
Figure 1.1. Original image and its gray scale segmented image 1
Figure 1.2. Stages in data mining 3
Figure 1.3. Diversity of clusters formation 8
Chapter 2.
Figure 2.1. Dataset used in this research 11
Figure 2.2. Stages in data mining 17
Figure 2.3. Data mining in medicine 20
Chapter 3.
Figure 3.1. Three well-separated areas of tumor formation 24
Figure 3.2. Variety of clusters 25
Figure 3.3(a) Creation of two clusters in brain data set 25
Figure 3.3(b) Creation of three clusters in brain data set 26
-- --
-- --
-- --
-- --
Chapter 6.
Figure 6.1. Performance calculation based on different parameters 37
Figure 6.2. Accuracy using Index A (a) no. of iteration=20 (b) no. of 39
iteration=10
Figure 6.3. Accuracy using Index B (a) no. of iteration=20 (b) no. of 40
iteration=10
Figure 6.4. Running time (in sec.) of all methods 46
 LIST OF TABLES

Page #
Chapter 1.
Table 1.1. Features of datasets 2
Table 1.2. Various formulae used 3
Chapter 2.
Table 2.1. Features of datasets 8
Table 2.2. Accuracy of method A 12
Table 2.3. Accuracy of method B 15
Chapter 3.
Table 3.1. Three well-separated areas of tumor formation 18
Table 3.2(a) Creation of two clusters in brain data set 20
Table 3.2(b) Creation of two clusters in brain data set 21
Table 3.3 Accuracy of datasets using Index ABC 24
-- --
-- --
-- --
-- --
-- --
Chapter 5.
Table 5.1. Performance calculation based on Index ABC 37
Table 5.2. Execution time for looping 39
Table 5.3. Assessment of different methods used 40
Table 5.4. Score value of all methods 46
 CONTENTS

CONTENT DETAILS PAGE NO.

Title Page i
Certificate ii
Declaration iii
Acknowledgements iv
Abstract v
List of Abbreviations vi
List of Figures vii
List of Tables viii
Contents ix

Chapter 1. Introduction 1 – 18

1.1. Introduction 3
1.2. Problem Statement 4
1.2. Proposed Method 4
1.2.1. Method to Improve the Performance of Discovery 5
1.2.2. Proposed Method II 6
1.3. Objectives of the Project 6
1.4. Organization of the Project 8
Summary 18

Chapter 2. Technology Used 19 – 29

2.1. Introduction 19
2.2. Method I 20
2.3. Method II 22
2.4. Method III 25
Summary 29

Chapter 3. Experimental Results 30 – 42

3.1. Introduction 30
3.2. Results and Discussion 31
3.2.1 Sub-sub-section 31
3.2.2 Sub-sub-section 33
3.2.3 Sub-sub-section 35
3.2.4 Sub-sub-section 38
3.3. Summary 42
Chapter 4. Conclusion and Future Scope 43 – 45

4.1. Future scope 44

4.2. Contributions 45

References 46 – 51

Appendix – A: List of Formulas Used I

Appendix – B: Snapshots III
CHAPTER 1
INTRODUCTION

The introduction chapter should contain the complete information of the project. By going
through the Introduction chapter, one can have a clear understanding of the contents of
other chapters. This chapter may include a discussion of the broad area of the work being
carried out which is followed by the sections viz, motivation, objective, scope of the
work, problem definition(s), etc. and the project organization i.e. a brief description of the
contents of other chapters. Finally, each chapter should conclude with a SUMMARY of
the things discussed.

1.1. BACKGROUND

In a conventional sense, image segmentation [5] [7] [10] is the partitioning of an image
into regions, where parts within a region are similar according to some uniformity
predicate, and dissimilar between neighbouring regions. Due to its importance, many
segmentation algorithms have been proposed, and a number of evaluation criteria [3] have
also been proposed. In spite of this, very few comparative studies on the methods used for
evaluation have been published [4] [6].

Typically, researchers show their segmentation results on a few images and point
out why the results ’look good’. We never know from such studies if the results are good
or typical examples. Since none of the proposed segmentation algorithms are generally
applicable to all images, and different algorithms are not equally suitable for a particular
application, there needs to be a way of comparing them, so that the better ones can be
selected. The majority of studies proposing and comparing segmentation methods
evaluate the results only with one evaluation method. However, results vary significantly
between different evaluators, because each evaluator may have distinct standards for
measuring the quality of the segmentation. An example demonstrating the concept of
image segmentation is shown in Figure 1.1. The figure shows an original image and its
corresponding gray-scale segmented image.
 .
Figure 1.1. Original image and its gray scale segmented image

1.2. PROBLEM STATEMENT

In this project the problem of computational optimization of discovery and selection of

semantic services is taken up with special focus to include disparate similarity demands of
clients and to provide flexibility in querying. The statement of the problem is explicitly
declared as:
“To computationally optimize discovery and selection of semantic services considering
similarity demands and flexibility in constructing queries using some alternate methods”.

1.3. OBJECTIVE AND MOTIVATION

“The objective of this project is to provide the reuse of third-party software components
which are becoming popular recently”.

Building systems based on reusable components is not a new idea. It has been proven to
be a very effective cost-reduction approach in the computer hardware industry. Today,
that industry is able to build computer systems based on standardized high-quality
hardware parts and devices reliably and quickly. Software engineers learned the value of
this idea many years ago.

We would like to introduce the issues and challenges in component testing approach:

a) Difficult to perform component analysis and testing due to the lack of the access to
component source code and internal artifacts.
b) Testing reused components in a new reuse context and environment.
c) Expensive cost of constructing component test bed, including test drivers and stubs.
In this work we proposed a model and an algorithm to generate test cases for component
composition. We take the help of UML state chart diagram and CIG for generating the
test cases.
1.4. PROPOSED METHOD

1.4.1. Method to Improve the Performance of Discovery

It is proposed to improve the performance of discovery using two techniques, namely,

clustering and indexing. These techniques achieve improved performance by eliminating
the irrelevant services of a query from semantic matching. Clustering and indexing are
complementary; clustering partitions services into groups of similar functional
characteristics and during discovery the cluster which is most similar to the query alone
will be chosen for semantic matching whereas indexing helps in retrieving services
mapped by the inputs and outputs of the query as can be seen from Figure 1.2.

Figure 1.2. Original image and its three clusters formation by applying proposed method

1.4.2. Evaluation Parameter ─ RMSE

RMSE [22] determines the difference in the results likely to be obtained by a model with
those which are actually present in it. RMSE is given by the following equation:-

M N
RMSE = 1/ M * N   ((x(i, j) - y(i, j)) 2
i=1 j=1
(1.1)

A smaller value obtained for RMSE implies the image is of premium quality.

The result of K-Means [3] was tested on different data sets taken from UCI repository [2].
The different features of the datasets are shown in Table 1.1 below.

Table 1.1. Features of datasets

Datasets No. of Attributes No. of Classes Instances

Iris 4 3 150
Wine 13 3 178
Seed 7 3 210
1.5. PROJECT ORGANIZATION

Chapter 1 titled, “Introduction”, presents the general overview of the concept of image
segmentation, techniques and applications involving it. It also contains the objective and
motivation behind the work.
Chapter 2 titled, “Literature review” presents the renowned works earlier performed by
well-known personalities in the area of image processing. This chapter furthermore
contains the positives and negatives of the existing works done.
Chapter 3 named as, “Proposed Method”,
Chapter 4 titled, “Experimental Results”
Chapter 5 titled, “Conclusion”, is the summery of the complete work carried out with a
miniature part given to the society.

SUMMARY
In this chapter we discussed about:
 The basic concept and goals of …
 Various applications where image segmentation plays a vital part.
 Parameters for evaluating the results like …, … and …
 Motivation, objective and domain of this work.
CHAPTER 2
TECHNOLOGY USED

This technology review is a search and evaluation of the available literature in your given
subject or chosen topic area. It documents the state of the art with respect to the subject or
topic you are writing about. It surveys the literature in your chosen area of study. If you
have to write a dissertation, you may be required to begin by writing a literature review.

The discussion has four main objectives:

a) It surveys the literature in your chosen area of study.
b) It synthesizes the information in that literature into a summary.
c) It critically analyses the information gathered by identifying gaps in current
knowledge; by showing limitations of theories and points of view; and by
formulating areas for further research and reviewing areas of controversy.
d) It presents the literature in an organized way.

………
………

In the end, conclude the chapter by including a SUMMARY of the things

discussed.

SUMMARY
In this chapter we discussed about:
 The basics of …
 The pros and cons of …
 The different models / techniques used by various …
REFERENCES

[1] A. Rajiv Kumar and S. Sonae. “Brain Tumor Detection using ANN”. Int. Conference
on Computing, pp. 270 – 275, 2015.

[2] C. Merz and P. Murphy. “UCI Repository of Machine Learning Databases”. Available:
fttp://ftp.ics.uci.edu/pub/machine-learning-databases.

[3] J. Mac Queen. “Some Methods for Classification and Analysis of Multivariate
Observations”. Fifth Berkeley Symposium on Mathematics, Statistics and Probability,
University of California Press, pp. 281 – 297, 1967.

[4] R. Gonzales, R. Woods and K. Reinhard. “Digital image processing”. Pearson

education international, New Jersey, pp. 301 – 328, 2008.

[5] T. M. Bag, D. B. Hollmask and M. R. Che. “Effectiveness of Pulsed Ultraviolet-Light

Treatment on Molases”. Journal of Biotechniques. vol. 5, issue 5, pp. 21 – 47, 2015.

[6] U. V. Lux. “Clustering stability”. Machine Learning and Artificial Intelligence, vol.
2, number 3, pp. 23 – 34, 2010.
 APPENDIX – A
LIST OF FORMULAS USED

K N
J =   d(Xi , C j )
j=1i=1
1. K-Means objective function :
M N
RMSE  1/ M * N  ((x(i, j)  y(i, j)) 2
i 1 j1
2. Root Mean Square Error :
 (Backside Hard Cover of the Project Report)

Department of Computer Science and Engineering

Silicon Institute of Technology,
Silicon Hills, Bhubaneswar –751024,
Odisha, India