ECE4553 Course Syllabus

Introduction to Pattern Recognition

4 ch (3C 2L)

Instructor Dr. Erik Scheme Term Fall 2024

Office RN Scott Hall, Room 202 Day Tues, Thurs
Phone (506) 458-7029 Time 8:30 AM – 9:50 PM
Email escheme@unb.ca Classroom D110 (Lab H117)
Website D2L, MS Teams Contact By email

Students are encouraged to contact me by email or Teams for assistance or to set up a meeting.

COURSE DESCRIPTION

ECE 4553 is an introductory course on pattern recognition and its applications. A wildly popular and
important field in industry and academia, pattern recognition is fundamental to human interaction and
is pervasive in today’s society. This course will introduce the basics steps in the design of a pattern
recognition system, including data acquisition, signal preprocessing, feature extraction, various
classification approaches, learning and adaptation, and system evaluation. Class sessions will be
conducted in lecture or workshop mode and employ case studies and hands-on exercises to reinforce
key ideas where possible. Topics will include Bayesian decision theory and parameter estimation,
feature selection and generation, parametric vs nonparametric classification, general machine learning,
supervised vs unsupervised learning, and clustering techniques.

Suggested (but not required) background: ECE3511, STATS 2593, familiarity with the following 1:
0F

 Linear Algebra, Matrix Analysis, Calculus

 Probability theory (Probability distribution, probability density, random variables)
 Matlab or Python (Assignments and labs are transitioning from Matlab to Python this year)

COURSE MATERIALS
There is a wide variety of online materials dedicated to machine learning and pattern recognition. There
are many excellent video tutorials on YouTube and free courses. Students are encouraged to leverage
these resources in whichever manner best suits their learning preferences.

Reference textbooks are purely optional, but include:

 Pattern Classification, 2nd Edition, by Duda, Hart, and Stork, Wiley 2001.
 Pattern Recognition, 4th Edition, by Theodoridis & Koutroumbas, Elsevier 2008
 Pattern Recognition and Machine Learning, by C. Bishop, Springer 2006.

1
The appendix in Duda, Hart, and Stork reviews much of the mathematics required.

1
A combination of email, MS Teams, and D2L will be used for communication of supporting material,
course content, and deliverables. Students are therefore expected to check each regularly (daily, at a
minimum) and are responsible for doing so. It is also advised that email is checked roughly an hour
before classes.

COURSE DELIVERY
This course is a combination of both lecture instruction and dedicated lab-time during which students
will advance their understanding of the discussed concepts. Lectures will introduce the principles of
pattern recognition theory, and when possible, draw upon examples and analogies to help explain how
each topic fits into the ‘big picture’, while lab-time will allow students to implement and experience
these lessons. Students may be asked to conduct their own reading ahead of classes to enable a more
interactive discussion.

MARKING SCHEME
The course will consist of lectures, assignments, classroom discussions, assigned readings, laboratories,
and a final project and report. Students are expected to participate in all course activities. Missed classes
and/or deliverables without previous arrangement may be subject to academic penalty.

Table 1: Breakdown of Grades

Assessment Value
Assignments 15%
Labs 20%
Progress Report 1 (Literature Review) 10%
Progress Report 2 (Preliminary Results) 15%
Final Project 30%
Presentations 10%
Total 100%

PROJECT
The term project will be completed in (nominal) groups of two. It will consist of the design and
assessment of a pattern recognition system for a particular application or experiment. A written
proposal for the topic, nature, and source of data for this project must be approved by the instructor.
The due date for this is to be determined and discussed in class. The project deliverables will consist of
two progress reports, a final journal publication style report (e.g. background, methods, analysis, and
conclusions), and a presentation at the end of the term. These presentations may be scheduled outside
of regular class times to allow for a fuller exploration and discussion. Although projects are marked as a
group, final grading will reflect observed disparities between the contributions of the group members.

Documents:
The final submission will consist of a (nominally) 7 to 8-page double-column IEEE style paper describing
the background, methods, and results of your work. You may use the IEEE Template for Conference
2
Papers.docx file uploaded to D2L. You should submit your document, along with any code and additional
figures or appendices to the dropbox on D2L. Formatting and deadlines for progress updates will be
discussed in class and/or templates will be provided on D2L.

Presentations:
You will also be required to present your progress and results to the class throughout and at the end of
the term. Details of the format and scheduling of these presentations will be discussed in class. You
should use a small number of thoughtfully designed PowerPoint slides to support your presentation
(one slide per minute is typically considered appropriate).

Peer Evaluation:
In addition to your document submission, you must also submit a detailed breakdown of the distribution
of work between the two group members. You may submit these individually through the separate
(individual) dropbox folder on D2L called “Peer Evaluation”. Details of this evaluation will be shared in
class.

ASSIGNMENTS
Assignments will be given to provide an opportunity to apply and test the knowledge learned in class.
Whereas labs will focus on implementing the content with real data, assignments will focus more on
theory, understanding, and design principles. Assignments are to be completed individually. Copying,
sharing, co-writing assignments, or similar, will not be tolerated and will be subject to academic penalty.

LABS
Labs will be completed individually. Lab reports will consist of a concise summary of the objectives, the
work completed during the lab, and value-added observations/learnings from its execution. Reports
should be formal (e.g. written in the third person) but concise and should focus on results and
demonstration of understanding. A tentative summary of labs and due dates is listed below but is
subject to change.

Table II: Expected Lab Dates

Lab # Lab Topic Date Due Date*

1 Introduction to Datasets 09/12 09/19
2 Data Visualization 09/26 10/03
3 Pre-Processing and Feature Extraction 10/10 10/17
4 Classification Performance Analysis 10/24 10/31
5 Eigenfaces and Facial Recognition 11/07 11/14
6 Unsupervised Learning 11/28 12/05
* Lab Due Dates: Lab reports are to be submitted via D2L by 5:00 pm, one week after the lab.

3
Services for Students with Disabilities
If you are a student with a disability of any type (physical, mental, learning, medical, chronic health,
sensory; visible or invisible) you are strongly encouraged to register with the UNBF Student Accessibility
Centre (SAC) (http://www.unb.ca/fredericton/studentservices/academics/accessibility/) so that you may
receive appropriate services and accommodations. Once you are registered with SAC, the instructor will
be notified via the UNBF SAC Accommodation Letter of your specific accommodations. If you would like
to discuss your particular needs with me, please book a time for a confidential appointment.

Student Services Writing and Study Skills Support

The UNB Writing and Study Skills Centre provides many coaching and mentoring services to assist with
writing papers, effective study methods, and other skills development related to student success:
http://www.unb.ca/fredericton/studentservices/academics/writing-centre/index.html

Student Services Learning Strategist Support

Any UNB student wanting to improve their academic skills may book appointments with the Learning
Strategist. The Strategist offers instruction on topics such as: “learning how to learn” strategies, memory
techniques, time management skills, test preparation and test-taking methods, note-taking, and other
learning and study skills.
https://www.unb.ca/fredericton/studentservices/academics/accessibility/learning-strategies.html

Equity, Diversity, and Inclusion

UNB embraces the idea of an intellectual community enriched by diversity along a number of
dimensions, including gender, gender identity, sexual orientation, age, culture, ability, race, ethnicity,
language, religion, and nationality. I intend that all students be well served by this course, that students'
learning needs be addressed both in and out of class, and that the diversity students bring to this class
be viewed as a resource, strength, and benefit. I intend to provide materials and activities that are
respectful of diversity. Your suggestions are encouraged and appreciated. Also, if any of our class
meetings conflict with your religious holidays, please let me know so that we can make arrangements
for you.

Office of Human Rights and Positive Environment: https://www.unb.ca/humanrights/index.html

Have a Concern, Comment or Suggestion?

As a student, you are responsible for putting in the effort required for success. If you are doing this,
however, and there is something that you feel could be done to improve the delivery of this material or
the overall course experience, please don’t hesitate to contact me. If you prefer anonymity, please feel
free to slip a note under my office door. Your feedback is appreciated!

ACADEMIC INTEGRITY
The University of New Brunswick places a high value on academic integrity and has a policy on
plagiarism, cheating, and other academic offenses.

Plagiarism includes:

1. quoting verbatim or almost verbatim from any source, including all electronic sources, without
acknowledgment;

4
 2. adopting someone else’s line of thought, argument, arrangement, or supporting evidence
without acknowledgment;
3. submitting someone else’s work, in whatever form without acknowledgment;
4. knowingly representing as one’s own work any idea of another.

NOTE: A penalty may be imposed on all members of a group unless an act of plagiarism is identified
clearly with an individual student or students.

Examples of other academic offenses include cheating on exams, tests, assignments or reports;
impersonating somebody at a test or exam; obtaining an exam, test, or other course materials through
theft, collusion, purchase, or another improper manner, submitting coursework that is identical or
substantially similar to work that has been submitted for another course; and more as set out in the
academic regulations found in the Undergraduate Calendar.

Penalties for plagiarism and other academic offenses range from a minimum of F (zero) in the
assignment, exam, or test to a maximum of suspension or expulsion from the University, plus a notation
of the academic offense on the student’s transcript.

For more information, please see the Undergraduate Calendar, Section B, Regulation VIII.A, or
visit http://nocheating.unb.ca. It is the student’s responsibility to know the regulations.