CSE 616 Applied Pattern Recognition

Taught: Winter 2001, Winter 2003, Winter 2005, Fall 2006

Oakland University, Michigan, USA
Oakland University, Michigan, USA


The CSE 616 course introduces basic ideas and concepts of object (or pattern) classification. We learn how to extract features from objects, reduce their cardinality without losing much information, represent these objects as feature vectors in an Euclidean space. We learn how to classify these feature vectors using statistical techniques such as parametric and non parametric density estimation. We finally explore Neural Networks (NN) techniques for pattern classification. We extend these isolated classification schemes to decode (or classify) a sequence of feature vectors using Hidden Markov Models (HMMs).Several applications selected by the students and the instructor are developed using C/C++ or Java programming languages. Some examples of applications are: speech recognition or identification, fingerprint/retina recognition, recognition of particular objects in images and other brand new applications. (Seminar on Pattern Recognition Online)



Duda et al, Pattern Classification, 2nd Edition with Computer Manual 2nd Edition Set, Wiley, May 2004.


Course Materials

Chapter 1
Chapter 2 (part 1)
Chapter 2 (part 2)
Chapter 2 (part 3)
Chapter 3 (part 1)
Chapter 3 (part 2)
Chapter 3 (part 3)
Chapter 4 (part 1)
Chapter 4 (part 2)
Chapter 5
Chapter 6
Chapter 10
MATLAB Session