ECE 759 601 Pattern Recognition

3 Credit Hours

Image pattern recognition techniques and computer-based methods for scene analysis, including discriminate functions, fixture extraction, classification strategies, clustering and discriminant analysis. Coverage of applications and current research results.

Prerequisite

ECE(CSC) 514 Random Processes, ST 371 Introduction to Probability and Distribution Theory. This is a course in Statistical Pattern Recognition. It relies heavily on a background in probability, as well as on a solid foundation in Linear Algebra. These topics are prerequisite, and will not be reviewed in this class.

Course Objectives

This is a course in Statistical Pattern Recognition. This course provides the quintessential tools to a practicing engineer faced with everyday signal processing classification and data mining problems. It heavily relies on a background in probability, as well as on a solid foundation in Linear Algebra. These topics are highly desirable to get the most out of the course.

Course Requirements

HOMEWORK: 10% The +/- or grade out of 10-system will be used for grading. Homework may be typed or handwritten.

FINAL EXAMINATION: 30% The Final Exam will be comprehensive but will focus on understanding rather than remembrance of results.

SOFTWARE REQUIREMENTS: Students are required to use computing facilities for writing programs for their homeworks. Although C/C++ programming is preferred, students are welcome to use Matlab and other mathematical packages available on departmental computers. All projects need to be written using word-processing software or typesetting tools.

PROJECTS: (2) Each worth 30%. Two projects will be assigned with a significant programming component and STRICTLY INDIVIDUAL WORK WILL BE EXPECTED!

The grading for this course is as follows:
Homeworks: 10% The +/- or grade out of 10-system will be used for grading.
Projects (2): 30% + 30%
Final Exam: 30%
The +/- or grade out of 10-system will be used for grading.

Required Textbook

Pattern Recognition (4th edition, 2014) by T. Sergios and K. Koutroumbas, ISBN 978-1-59749-272-0, Elsevier Science and Technology.

Other notes may at times be used.