2009 Spring

Pattern Recognition

Instructor: Shih-Hung Wu

 

課號

課名

開課班級

上課時間地點

7793

圖形識別 

Pattern Recognition

資訊工程系碩士在職專班1A  

Wed A-C M-312

The goal of this course is to provide the students with a basic knowledge of pattern recognition. The students will realize the following concepts in the course:
1.Classifiers based on Bayes decision theory
2.Linear/nonlinear classifiers
3.Feature selection
4.Feature generation
5.Context-dependent classification
6.System evaluation
7.Clustering algorithms 

 

Textbook:

Christopher M. Bishop, "Pattern Recognition and Machine Learning", Springer, 2006.

References

Pattern Recognition (2nd ed.) by S. Theodoridis and K. Koutroumbas
Academic Press
全華圖書

Pattern Classification (2nd ed.)
by Richard O. Duda, Peter E. Hart and David G. Stork
Wiley Interscience
680 pages
ISBN: 0-471-05669-3
歐亞書局

投影片:

Bishop: Official slides

    Theodoridis: Official slides

    Duda: Ch1, Ch2_1, Ch2_2, Ch2_3, Ch3_1, Ch3_2, Ch3_3, Ch4_1, Ch4_2, Ch2.11 Bayesian Network, Ch5, SVM, Ch6, CH10

Schedule

Week

Date

Event

 

Presentation list

Notes

1

2/18

 

Introduction

X

 

2

2/25

 

Bayesian Rule

X

 

3

3/4

 

Bayesian Decision, Normal Distribution

X

 

4

3/11

 

 

X

 

5

3/18

 

Maximum Likelihood

 

 

6

3/25

 

Maximum Likelihood

 

 

7

4/1

 

Hidden Markov Model

 

 

8

4/8

 

Hidden Markov Model

 

 

9

4/15

midterm

 

X

Open Book test

10

4/22

 

K-NN

 

 

11

4/29

 

Neural Network

 

 

12

5/6

 

Support Vector Machine

 

 

13

5/13

 

Template matching

 

 

14

5/20

 

Clustering

 

 

15

6/3

 

AdaBoost

 

 

16

6/10

Term Project Demo

 

期末專題口頭報告(準備投影片DEMO)

 

17

6/17

Term Project Demo

 

期末專題口頭報告(準備投影片DEMO)

 

18

6/24

Final Exam.

 

期末專題繳交書面報告

 

Homework

#1. Term Project Report Proposal (4/1 due)

#1. Term Project Report Draft (5/6 due)

Brief Presentation

20 min presentation and 5 min Q & A.

Term Project Report Presentation

10 min presentation and 5 min Q & A.

Term Project Report

Cover page, Introduction, model, dataset, system, discussion.

1. problem definition

2. Collect data

3. data -> feature vector   (X1, X2, ... , Xn) Class tag

4. train classifier

5. test

 


Last update: 2006/10/02

shwu@cyut.edu.tw