Image Processing, Comp.

Vision & Pattern Recognition Group In-Course Page 1 of 9

IMAGE PROCESSING, COMPUTER VISION AND PATTERN RECOGNITION

(CT036-3-3-IPPR)
IN–COURSE ASSESSMENT
(GROUP PROJECT)

Intake : APT3F1706CYB
Lecturer : MD BAHARUL ISLAM
Email: baharulislam.md@apu.edu.my
Date assigned : Week 2
Due date : Week 14 (Strictly follow this deadline)

1.0 THE PROJECT OVERVIEW

You have an option to choice ONE of the following projects:
HSVS: Hand Signature Verification System

Due to the security issue, an automatic hand signature verification system is very
important, especially for financial institutes like Bank. The traditional verification
system is laborious and time consuming. To prevent the fraud case, the automatic
signature verification system can play a significant role. You have to develop an
application/system that can automatically verify your/any hand signature.
Option 1

HDC: Human Detection and Counting

Level 3 Asia Pacific University of Technology and Innovation Nov 2017

Image Processing, Comp. Vision & Pattern Recognition Group In-Course Page 2 of 9

Driven by the significant advancements in technology and social issues such as

security management, there is a strong need to detect human motion and counting
them for a certain scenario. You know that Technology Park Malaysia is a very busy
area, especially in working days. Is it possible to detect and count number of people
enter the human enters to TPM/day? Develop a system that can enable to detect the
human and counting them for a given video/live video. The given offline video can be
the CCTV footage (e.g. shopping mall, TPM, or any public dataset). Otherwise, its can
be live video while human can move throughout the scene.
Option 2

2.0 OBJECTIVES OF THIS PROJECT:

Overall objectives: Equip students with practical knowledge about image processing, computer
vision and pattern recognition in programming and computing. The focus will not be on teaching
all of the details of programming under specific platforms, but rather on providing both a high-level
understanding and practical implementation experience of reusable algorithms and coding
techniques that apply to the development of applications across different platforms and genres.

Project specific objectives:

Option 1-HSVS: Provide an automatic solution to verify the hand signature for preventing the
fraud case, especially for financial institutions.

Level 3 Asia Pacific University of Technology and Innovation Nov 2017

Image Processing, Comp. Vision & Pattern Recognition Group In-Course Page 3 of 9

Option 2-HDC: Provide a solution to automatically detect the human motion and counting total
number of human in certain time. In this application, you will enable to count all the people who
visited TPM/APU.

3.0 LEARNING OUTCOMES:

At the end of this project, the students should be able to:
 Demonstrate a working knowledge and understanding of image processing and computer
vision algorithms.
 Use different approaches to edit/extract and manipulate image using the existing algorithms.
 Identify and adapt appropriate algorithms and software to judge resource requirements for
an imaging problem.
 Design and build appropriate systems for problems requiring an imaging solution

4.0 PROJECT REQUIREMENTS:

The given assignment must satisfy the following requirements:

HSVS: Hand Signature Verification System

A set of signatures is given that can be used as training dataset. Your goal is to
verify the given signature automatically as approved/reject status.
Option 1:
 Minimum 5 signatures have to verify in the document;
Requirements
 The signature alphabets/numbers can be in English, not any other
languages;
 The signature can be offline
HDC: Human Detection and Counting

Option 2: A set of sample videos is given. The goal of your application is to detect the
Requirements human in the given video and then, count all of them.

 Your system may work also on live video (e.g. Surveillance video)
 No need to track any other objects like car/bus

Level 3 Asia Pacific University of Technology and Innovation Nov 2017

Image Processing, Comp. Vision & Pattern Recognition Group In-Course Page 4 of 9

5.0 PROJECT MEMBER:

Group work: 2-3 members / group, preferably 2 students/group.

6.0 DELIVERABLES:
The implemented application can be delivered in the form of a CD-ROM and the documentation
(project report in printed form).

6.1 APPLICATION IN CD FORM:

 The completed application must be compiled into an executable file (*.exe) and
burned onto a CD-ROM.
 The CD-ROM must contain all relevant source code (*.cs, *.cpp, *.m or *.h,) and
test files (e.g. *.jpg, *.png etc.) along with the soft copy of your report.
 Do not compress your source code.
 Please write a ReadMe file where you should highlight the instructions of your
application.

6.2 DOCUMENTS: PROJECT REPORT

As part of your assessment, you will have to submit a project report which includes the
followings:
 Table of contents
 Contribution matrix
 Acknowledgement
 Abstract: The objectives of this project (200-300 words)
 Introduction: Analysis the problem
 Description and justification of the proposed algorithms
 Experimental Results
 Description and discussion on the obtained results
 Critical comments, analysis and future work direction
 Conclusion

Level 3 Asia Pacific University of Technology and Innovation Nov 2017

Image Processing, Comp. Vision & Pattern Recognition Group In-Course Page 5 of 9

 List of References (Scholarly references, not wiki, personal websites, blogs,

forums)

Take a priority note:

 Do not include programming codes in your documentation. You may use the algorithm that you
followed.
 The total number of recommended pages for your documentation/report is in the range between
20-30 pages;
 You may use different sources of information, including internet and previous related works. If
you have accessed the internet information, please give credit to them by citation and references
throughout the document.
 All references must be made using the Harvard style as shown below:
The theory was first propounded in 2001 (Larsen, A.E. 2001), but since then has been refuted;
M.K. Larsen (1983) is among those most energetic in their opposition……….

List of references at the end of your document must be specified in the following format:

1. Santella, A., Agrawala, M., DeCarlo, D., Salesin, D. and Cohen, M., 2006, April. Gaze-based
interaction for semi-automatic photo cropping. In Proceedings of the SIGCHI conference on Human
Factors in computing systems (pp. 771-780). ACM.

2. Liu, L., Chen, R., Wolf, L. and Cohen‐Or, D., 2010, May. Optimizing photo composition.
In Computer Graphics Forum (Vol. 29, No. 2, pp. 469-478). Blackwell Publishing Ltd.

3. Gonzalez, R.C. and Wintz, P., 1977. Digital image processing. McGraw-Hill, London.

4. Larsen, M.K. 1983, British Medical Journal (Online), Available World Wide Web: URL:
http://libinfor.ume.maine.edu/acquatic.htm (Accessed 19 November 2015)

7.0 ASSESSMENT CRITERIA

The project will be assessed based on criteria, shown in Table 1. Please remind these
criteria during your project development and documentation.

Level 3 Asia Pacific University of Technology and Innovation Nov 2017

Image Processing, Comp. Vision & Pattern Recognition Group In-Course Page 6 of 9

Table 1: Assessment criteria

Mid-Point Evaluation (10%) Presentation (10%)
Idea and working plan, initial 10% Content, delivery style, 10%
workflow professional style, good
use of language.
Documentation (40%)* Implementation (40%)*
Criteria Marks Criteria Marks Allocated
Allocated
Objectives 5% Performance results of 20%
Problem specification 5% the algorithms

Description and justification of the 10%

proposed algorithms.

Description and discussion of the 10% Technicality:

test plan and results obtained Implementation of 20%
algorithms
Critical comments and analysis 10%

* Each group member should have minimum 30% individual contribution in their group
assignment. You will be asked during your presentation. The midpoint presentation (progress
observation) will be held at WEEK 8/9. The group in-course final presentation will be held at
WEEK 15/16.

8.0 FRONT COVER OF PROJECT:

All reports must be prepared with a front cover. A protective transparent plastic sheet can be
placed in front of the report to protect the front cover. The front cover should be presented with
the following details:

a) Names.
b) Intake code.
c) Subject.
d) Project Title.
e) Date Completed (the report is due to be handed in).

Level 3 Asia Pacific University of Technology and Innovation Nov 2017

Image Processing, Comp. Vision & Pattern Recognition Group In-Course Page 7 of 9

9.0 DEVELOPMENT TOOLS

The code of this assignment may be written in any preferable programming language,
shown as Table 2. My recommendation to the students is to utilize the MATLAB/OCTAVE
image processing and computer vision toolboxes or packages. You easily access some user
friendly built-in library that can help you in different ways to improve your programming
capability during the development of this assignment.

Table 2: Examples of the software/tools to use in this assignment

Tools / Software
1 C Programming
3 Microsoft Visual C++ (MFC Programming)
4 MATLAB (recommended)
5 OCTAVE (recommended)

 The program must contain all the relevant source code. The usage of predefined
imaging functions in the image processing library is allowed. However, you must
understand fully how the codes are used because this will be tested in the
presentation. You must also identify through comments in your code which code
has been reused, citing and referencing it appropriately.
 The writing of code and the work behind it must belong substantially to you. If any
person besides you contributes in any significant way to the assignment, you must
credit their work in your description. Similarly, if you include information/source
code that you have used from other published sources or web pages, you must cite
them as references. Otherwise, the content will be counted as plagiarism.

10.0 ACADEMIC INTEGRITY

 You are expected to maintain the utmost level of academic integrity during the duration of the
course.
 Plagiarism is a serious offense and will be dealt with according to APIIT’s regulations on
plagiarism.

Level 3 Asia Pacific University of Technology and Innovation Nov 2017

Image Processing, Comp. Vision & Pattern Recognition Group In-Course Page 8 of 9

11.0 WORKLOAD MATRIX:

All members are required to sign off your workload matrix, before submitting your
assignment.

12.0 PERFORMACE CRITERIA

DISTINCTION
 Demonstrate a critical approach in proposing the solution.
 Demonstrate innovative and creative approach in creating the final application
prototype.
 Excellent/relevant research into application area.
 Excellent report with substantial content, clearly distinguishing individual contribution
from group ones.
 Application prototype demonstrates considerable flare, works well and was
demonstrated successfully.

CREDIT
 Demonstrate a critical approach in proposing the solution.
 Demonstrate sound good ability in creating the final application prototype.
 Good research into application area.
 Report well-structured and content good. Application prototype works well and was
demonstrated successfully.

PASS
 Solution proposed not always appropriate and lacks a critical approach.
 Demonstrate poor workable final application prototype.
 Poor research on the application.
 Report somewhat confused, lacking in examples and several areas not fully explained.
 Application prototype not fully working and demonstration not very successful.

MARGINAL FAIL
 Solution proposed not appropriate and lacks a critical approach.
 Research into application area insufficient.
 Report (if present) confused, lacking in examples and several areas not fully explained.
 Application prototype not working and did not contribute much to the demonstration.

FAIL
 Fails to justify the solution.
 Very poor on report writing.
 Report (if present) far too brief.
 No evidence of any application prototype produced

Level 3 Asia Pacific University of Technology and Innovation Nov 2017

Image Processing, Comp. Vision & Pattern Recognition Group In-Course Page 9 of 9

 Product not working and did not contribute to the demonstration

Level 3 Asia Pacific University of Technology and Innovation Nov 2017