S004102-模式识别

发布者:沈如达发布时间:2018-04-23浏览次数:16

研究生课程开设申请表

 开课院(系、所): 信息科学与工程学院

 课程申请开设类型: 新开□     重开    更名□请在内打勾,下同

课程

名称

中文

模式识别

英文

Pattern Recognition

待分配课程编号

S004102

课程适用学位级别

博士


硕士

总学时

32

课内学时

32

学分

2

实践环节


用机小时


课程类别

公共基础     专业基础      专业必修     专业选修

开课院()

信息科学与工程学院

开课学期

秋季

考核方式

A.笔试(开卷     闭卷)      B. 口试    

C.笔试与口试结合                 D. □其他

课程负责人

教师

姓名

秦中元

职称

副教授

e-mail

zyqin@seu.edu.cn

网页地址


授课语言

汉语

课件地址


适用学科范围

二级学科

所属一级学科名称

信息与通信工程

实验(案例)个数


先修课程

线性代数、概率论与随机过程和计算机程序设计语言

教学用书

教材名称

教材编者

出版社

出版年月

版次

主要教材

模式分类

RODuda等著,李宏东等译

机械工业出版社

2005

第一版

主要参考书

模式识别

蔡元龙

西安电子科技大学出版社

1992

第一版

模式识别——原理、方法及应用

J.P.Marques de Sa著,吴逸飞译

清华大学出版社

2002

第一版

模式识别

边肇祺,张学工

清华大学出版社

2000

第一版


一、课程介绍(含教学目标、教学要求等)300字以内)

本课程是信息安全专业的专业必修课。模式识别是一门理论与应用并重的技术科学,其目的是利用计算机对某些物理现象进行分类,在错误概率最小的条件下,使识别的结果尽量与事物相符。模式识别的原理和方法在医学、军事等众多领域应用十分广泛,是计算机及其相关专业进行科学研究的基础。本课程主要介绍了统计模式识别中的基本理论,包括聚类分析、贝叶斯决策理论、判别函数、特征提取和选择及相关应用。通过本课程的学习,同学在了解了上述经典方法的基本原理的基础上,要能够阅读相关中外文献,了解如模糊数学、神经网络等新技术在模式识别领域的应用,了解模式识别的发展趋势及其有效应用。


二、教学大纲(含章节目录):(可附页)

第一章 概论

1.1模式识别的概念

1.2模式识别系统

1.3模式识别的应用

1.4特征矢量和特征空间

1.5随机矢量的描述

1.6多维正态分布

第二章 聚类分析

2.1 模式相似度测度

2.2 聚类准则

2.3 基于试探的聚类搜索算法

2.4 层次聚类法

2.5 动态聚类法

2.6 聚类有效性

第三章 贝叶斯决策理论

3.1 基于最小错误率的贝叶斯决策

3.2 基于最小风险的贝叶斯决策

3.3 最小最大决策.

3.4 贝叶斯分类器和判别函数

3.5 正态分布时的贝叶斯决策法则

第四章  线性判别函数

4.1 线性判别函数

4.2 判别函数值的鉴别意义、权空间及解空间

4.3 Fisher线性判别

4.4 感知器算法及梯度下降法

4.5 广义线性判别函数

4.6 支持向量机

第五章   特征提取与选择

5.1 概述

5.2 类别可分性判据

5.3 基于可分性判据进行变换的特征提取与选择

5.4 按概率距离判据的特征提取方法

5.5 基于熵函数的可分性判据

5.6 K-L变换及其在特征提取与选择中的应用


第六章  神经网络识别理论及模型

6.1 人工神经网络基本模型

6.2 神经网络分类器

6.3 Hopfield模型

6.4 前馈神经网络及其主要方法


三、教学周历

 周次

 教学内容

 教学方式

1

概论

讲课

2

聚类分析

讲课

3

聚类分析

讲课

4

特征提取与选择

讲课

5

特征提取与选择

讲课

6

贝叶斯决策理论

讲课

7

贝叶斯决策理论

讲课

8

线性判别函数

讲课

9

线性判别函数

讲课

10

神经网络识别理论及模型

讲课

11

神经网络识别理论及模型

讲课










四、主讲教师简介:

秦中元,男,博士,副教授,硕士生导师。19969月于西安交通大学计算机科学与工程系软件专业获学士学位;19993月于西安交通大学电子与信息工程学院获硕士学位;200311月毕业于西安交通大学电信学院图像处理与模式识别研究所,获通信与信息系统专业工学博士学位。同年12月进入东南大学无线电系信息安全研究中心任教。在电子学报等重要期刊和IEEE学术会议上发表论文十余篇。目前主要研究兴趣为:模式识别与智能视频监控、无线网络安全。


五、任课教师信息(包括主讲教师):

 任课

 教师

 学科

 (专业)

 办公

 电话

 住宅

 电话

 手机

 电子邮件

通讯地址

 邮政

 编码

秦中元

信息安全

83795112-801

83790986

13951031560

zyqin@seu.edu.cn


210096





Application Form For Opening Graduate Courses

School (Department/Institute)School of Information Science and Engineering

Course Type: New Open □   Reopen      Rename □Please tick in □, the same below

Course Name

Chinese

模式识别

English

Pattern Recognition

Course Number

S004102

Type of Degree

Ph. D


Master

Total Credit Hours

32

In Class Credit Hours

32

Credit

2

Practice


Computer-using Hours


Course Type

Public Fundamental    □Major Fundamental    Major Compulsory     □Major Elective

School (Department)


Term

Autumn

Examination

A.PaperOpen-book    Closed-bookB. □Oral   

C. □Paper-oral Combination                       D. □ Others

Chief

Lecturer

Name

Qin Zhongyuan

Professional Title

Associate Professor

E-mail

zyqin@seu.edu.cn

Website


Teaching Language used in Course

Chinese

Teaching Material Website


Applicable Range of Discipline

second-class discipline

Name of First-Class Discipline

Information and Communication Engineering

Number of Experiment


Preliminary Courses

Linear AlgebraicProbability TheoryComputer Foundation

Teaching Books

Textbook Title

Author

Publisher

Year of Publication

Edition Number

Main Textbook

Pattern Classification

R.C.Duba, etc

China Machine Press

2005

1

Main Reference Books

Pattern Recognition

Cai Yuanlong

XiDian University Press

1992

1

Pattern Recognition - Concepts, Methods and Applications

J.P.Marques de Sa

Tsinghua University Press

2002

1

Pattern Recognition

Bian Zhaoqi, Zhang Xuegong

Tsinghua University Press

2000

1


  1. Course Introduction (including teaching goals and requirements) within 300 words:

Pattern Recognition is the scientific discipline whose goal is the classification of objects into a number of categories or classes with the minimum error probability. Depending on the application, these objects can be images or signal waveforms or any type of measurements that need to be classified. Its principle and methods are widely used in many fields, such as medicine, military, etc. and it is the basis of scientific research of computer science and relevant major.

The basic theories of statisticalpattern recognition are introduced in this course, including cluster analysis, Bayesian decision theory, discriminant function, feature extraction and selection and their applications. The neural network and fuzzy mathematics in pattern recognition are also introduced.

After the study of this course, the basic concepts and principles of pattern recognition should be mastered. The students also should can read the newest materials and learn the development trends of pattern recognition and its applications.


  1. Teaching Syllabus (including the content of chapters and sections. A sheet can be attached):

Ch1. Introduction

1.1 Concept of Pattern Recognition

1.2 Pattern Recognition System

1.3 Applications of Pattern Recognition

1.4 Feature Vectors and Feature Space

1.5 Description of Random Vectors

1.6 Multi-dimensional Normal Distribution


Ch2. Cluster Analysis

2.1 Measurement of Pattern Proximity

2.2 Clustering Principles

2.3 Clustering Algorithms Based on Test

2.4 Hierarchical Clustering Algorithms

2.5 Dynamic Clustering Algorithms

2.6 Cluster Validity


Ch3. Bayesian Decision Theory

3.1 Bayesian Decision Base on Minimum Error Probabilities

3.2 Bayesian Decision Base on Minimum Risk

3.3 Minmax Decision

3.4 Bayesian Classifier and Discriminant Functions

3.5 Bayesian Classification for Normal Distributions


Ch4.Linear Discriminant Functions

4.1 Linear Discriminant Functions

4.2 Meaning of Discriminant Function Values, Weight space and Solution Space

4.3 Fisher Linear Discriminant

4.4 Perception and Gradient Descendant Method

4.5 Generalized Linear Discriminant Functions

4.6 Support Vector Machines


Ch5. Feature Extraction and Selection

5.1 Introduction

5.2 Class Separability Criterion

5.3 Feature Extraction and Selection Based on Separability Criterion

5.4 Feature Extraction According to Probability Distance

5.5 Separability Criterion Base on Entropy Functions

5.6 K-L Transform and its use in Feature Extraction and Selection


Ch6. Neural Network Theory and its Model

6.1 Basic Model of Neural Network

6.2 Neural Network Classifier

6.3 Hopfield Model

6.4 Feed-forward Neural Network and its main methods


  1. Teaching Schedule:

Week

Course Contents

Teaching Method

1

Introduction

lecture

2

Cluster Analysis

lecture

3

Cluster Analysis

lecture

4

Bayesian Decision Theory

lecture

5

Bayesian Decision Theory

lecture

6

Discriminant Functions and Trained Decision Classifier

lecture

7

Discriminant Functions and Trained Decision Classifier

lecture

8

Feature Extraction and Selection

lecture

9

Feature Extraction and Selection

lecture

10

Neural Network Theory and its Model

lecture

11

Neural Network Theory and its Model

lecture




18




Note: 1.Above one, two, and three items are used as teaching Syllabus in Chinese and announced on the Chinese website of Graduate School. The four and five items are preserved in Graduate School.

2. Course terms: Spring, Autumn , and Spring-Autumn term.  

3. The teaching languages for courses: Chinese, English or Chinese-English.

4. Applicable range of discipline: public, first-class discipline, second-class discipline, and third-class discipline.

5. Practice includes: experiment, investigation, research report, etc.

6. Teaching methods: lecture, seminar, practice, etc.

7. Examination for degree courses must be in paper.

8. Teaching material websites are those which have already been announced.

9. Brief introduction of chief lecturer should include: personal information (date of birth, gender, degree achieved, professional title), research direction, teaching and research achievements. (within 100-500 words)


  1. Brief Introduction of Chief lecturer:

Qin Zhongyuan was born in Jan. 1974 in Anyang, Henan province, China. He received his B.E. and M.E. degree both in Computer Science from Xi'an Jiaotong University in 1996 and 1999, respectively, and in Nov. 2003 got his Ph.D. degree in Communication and Information Systems in the Electronic and Information Engineering Institute of Xi'an Jiaotong University. He is now an associate professor in the Research Center of Information Security of Southeast University. His main research interests include pattern recognition, digital video processing, security in wireless network,etc.


  1. Lecturer Information (include chief lecturer)


Lecturer

Discipline

(major)

Office

Phone Number

Home Phone Number

Mobile Phone Number

Email

Address

Postcode

Qin Zhongyuan

Information Security

83795112-801


13951031560

zyqin@seu.edu.cn


210096







9