1
课程大纲
COURSE SYLLABUS
1.
课程名称(中英文)
Course Title(Chinese
and English)
模式识别 Pattern Recognition
m2.
课程类别 Course Type
专业课
3.
授课院系
Originating Department
电子与电气工程系 Department of Electronic and Electrical Engineering
4．
可选课学生所属院系
Open to Which Majors
本专业
5.
课程学时 Credit Hours
64
6.
课程学分 Credit Value
3
7.
授课语言
Teaching Language
中英文
8.
授课教师 Instructor(s)
（如果是一个课题组共同讲
授的，请标明 MI 以及其他构
成成员。）
时红建
9.
先修课程、其它学习要求
Pre-requisites or Other
Academic Requirements
信号与系统(Signal and System)
10.
教学目标 Course Objectives
课程学习目标：到课程结束，学生能做如下：
推导与程序执行贝页斯理论
进行参数与非参数密度分布估计
对分离与不可分离的模式，涉及分类器
使用神经网络进行分类
Course Learning Objectives: By the end of this class, students would:
Derive and implement Bayes Decision Theory.
Perform Parametric and Non Parametric Density Estimation.
Design Linear Classifiers for separable and non-separable patterns.
Implement classifiers using neural networks.
2
11.
教学方法及授课创新点 Teaching Methods and Innovations
Fundamentals of Statistical, Structural, and Neural Pattern Recognition Approaches:
Parametric and Nonparametric Classification, Feature Extraction, Clustering, Self-organizing Nets
for Pattern Recognition, and Formal Languages Representation, Current medical and industrial
applications.
12.
教学内容及学时分配 Course Contents and Course Schedule
COURSE OUTLINE WEEKS
1. INTRODUCTION ...................................................................................................……………1
1.1 MATHEMATICAL FOUNDATION
2. BAYES DECISION THEORY..............................................................................…………………..2
2.1 Bayes Classier for Continuous Case
2.2 The Gaussian Two-class classifier
2.3 Bayes Classier for Discrete Case
2.4 Error Probability and Receiver Operating Characteristics
3. MAXIMUM-LIKELIHOOD AND BAYESIAN PARAMETER ESTIMATION.................................2
3.1 Maximum Likelihood Estimation
3.2 Application to Bayesian Classification
3.3 Learning the Mean of Gaussian Density Function
4. NONPARAMETRIC TECHNIQUES ........................................................................................3
4.1 Probability Density Estimation
4.2 Parzen Windows Estimation
4.3 k Nearest Neighbor Estimation
4.4 Nearest Neighbor Rule
4.5 k Nearest Neighbor Rule
5. LINEAR DISCRIMINANT FUNCTIONS...................................................................................2
5.1 Linear Discriminant Functions and Decision Surfaces
5.2 The Two-Category Case
5.3 Generalized Linear Discriminant Functions
5.4 Relaxation Procedure
5.5 Minimum Square Error Procedure
5.6 Linear Programming Procedure
13.
课程考核 Course Assessment
Homework (30%), Projects (40 %), and three Exams (30%).
14.
教材及其它参考资料 Textbook and Supplementary Readings
R. O. Duda, P. E. Hart and Stork, Pattern Classification and Scene Analysis, Wiley, New York, 2nd
Edition 2001.
3
