Syllabus  |   Lectures  |   Downloads  |   FAQ  |   Ask a question  |  
Course Co-ordinated by IIT Kharagpur
Coordinators
 
Prof. P.K. Biswas
IIT Kharagpur

 

Download Syllabus in PDF format



Untitled Document

The course has been designed to be offered as an elective to final year under graduate students mainly from Electrical Sciences background. The course syllabus assumes basic knowledge of Signal Processing, Probability Theory and Graph Theory. The course will also be of interest to researchers working in the areas of Machine Vision, Speech Recognition, Speaker Identification, Process Identification etc.
The course covers feature extraction techniques and representation of patterns in feature space. Measure of similarity between two patterns. Statistical, nonparametric and neural network techniques for pattern recognition have been discussed in this course. Techniques for recognition of time varying patterns have also been covered. Numerous examples from machine vision, speech recognition and movement recognition have been discussed as applications. Unsupervised classification or clustering techniques have also been addressed in this course.
Analytical aspects have been adequately stressed so that on completion of the course the students can apply the concepts learnt in real life problems.

 
 

S. No.

Topics

No. of Lectures

1.

Introduction 1 Lecture
Feature extraction and Pattern Representation
Concept of Supervised and Unsupervised Classification
Introduction to Application Areas

1

2.

Statistical Pattern Recognition
Bayes Decision Theory
Minimum Error and Minimum Risk Classifiers
Discriminant Function and Decision Boundary
Normal Density
Discriminant Function for Discrete Features
            Parameter Estimation

7

3.

Dimensionality Problem
Dimension and accuracy
Computational Complexity
Dimensionality Reduction
Fisher Linear Discriminant
Multiple Discriminant Analysis

5

4.

Nonparametric Pattern Classification
Density Estimation
Nearest Neighbour Rule
            Fuzzy Classification

5

5.

Linear Discriminant Functions
Separability
Two Category and Multi Category Classification
Linear Discriminators
Perceptron Criterion
Relaxation Procedure
Minimum Square Error Criterion
Widrow-Hoff Procedure
Ho-Kashyap Procedure
             Kesler’s Construction

7

6.

Neural Network Classifier
Single and Multilayer Perceptron
Back Propagation Learning
Hopfield Network
Fuzzy Neural Network

7

7.

Time Varying Pattern Recognition
First Order Hidden Markov Model
Evaluation
Decoding
Learning

4

8.

Unsupervised Classification
Clustering
Hierarchical Clustering
Graph Based Method
Sum of Squared Error Technique
Iterative Optimization

4

 

Total

40

Preliminary knowledge of Probability Theory, Signal Processing and Graph Theory.


  1. Richard O. Duda, Peter E. Hart and David G. Stork, "Pattern Classification", John Wiley & Sons, 2001.
  2. Earl Gose, Richard Johsonbaugh and Steve Jost, "Pattern Recognition and Image Analysis", Prentice Hall, 1999.


Important: Please enable javascript in your browser and download Adobe Flash player to view this site
Site Maintained by Web Studio, IIT Madras. Contact Webmaster: nptel@iitm.ac.in