|
Applied Pattern Recognition
CSE 616 Fall 2006 (4 cr),
Prerequisites: CSE 513 Monday
& Wednesday, from 5:30 to 7:17 pm, Room 187 SEB
Instructor
Djamel Bouchaffra, PhD
Assistant Professor of Computer
Science
Office: 131 Dodge Hall
Phone: (248) 370-2242
Fax: (248) 370-4625
Office Hours: Monday
& Wednesday from 4:15 to 5:15 pm
Description
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)
|
