《计算机视觉》课程简介

Introduction of Computer Vision

 

课程名称 (Course Name) ：计算机视觉Computer Vision

课程代码 (Course Code）：C032703

学分/学时 (Credits/Credit Hours)：3.0 /54

开课时间 (Course Term )：春 spring

开课学院（Course School）: 电子信息与电气工程学院  seiee

任课教师（Teacher）: 刘允才  liu yuncai

面向专业（Specialized Field）: 模式识别与智能控制  pattern recognition and intelligent system

课程讨论时数（Course Hours）:  4  小时(Hours)

课程实验数（Lab Hours）:    8  小时(Hours)

 

课程内容简介(Course Contents Introduction):

Computer vision aims to recover useful information about a 3D scene from its 2D projections (images), such as the depth and structure, motion, surfaces curvature and orientation of 3D objects and status and meaning of the actions of 3D scene. In this course, basic concept, theories and algorithms of computer vision are introduced. First, basic operations of image will be reviewed, then relative theories and algorithms of regions, edge detection, stereo vision, 3D motion analysis, contour, texture, shading, optical flow, camera calibration, curves and surfaces of reconstructed 3D objects, dynamic vision systems will be discussed in details.

 

教学大纲(Course Outline):

第一讲  绪论 (Introduction)

1. Concept of computer vision

2. Related fields

3. Perspective Projection

4. Digital Image

第二讲  图像滤波 (Image Filtering)

1. Linear system

2. Mean Filter

3. Gaussian Filter

4. Discrete Guassian Filter

5. Median Filter (nonlinear)

6. Other nonlinear filters

7. Morphological Filters

8. Histogram Modification

第三讲  二进制图像处理 (Binary Image Processing)

1. Neighbors/path/connectivity

2. Geometric attributes of a binary image/component

3. Binary Processing

第四讲  区域 (Region)

1. Region Representation

2. Region Segmentation

第五讲  边沿检测 (Edge Detection)

1. Edge detection

2. Gradient Based Edge Detection

3. LapLacian Operator

4. LapLacian of Gaussian

5. Canny Edge Detector

6. Corner Detection

7. Line Detection

第六讲  立体视觉 (Stereo)

1. Depth and stereo

2. Typical stereo imaging system

3. Stereo matching

4. Stereo of arbitrary camera arrangement

5. Structured Lighting

第七讲  运动的理解与估值 (Motion)

1. 3D motion expression

2. Motion from 3D PCs

3. Motion from 2D PC

4. Motion from LC’s

5. Motion from other image clues

第八讲  轮廓 (Contours)

1. Digital Curve

2. Curve Reporesentations

3. Curve fitting

4. Regression

5. Hough Transform

第九讲  纹理 (Texture)

1. Gray-level co-occurrence Matrix

2. Structural analysis of texture

3. Model-based Texture Analysis

4. Shape From Texture

第十讲  图像光度学 (Shading)

1. Basic Concepts

2. Surface Reflectance

3. Shape from shading

4. Photometric stereo

第十一讲  光流场 (Optic Flows)

1. Concept

2. Constraint equation of optical flow

3. Solve for optical flow

4. Understand optical flow

第十二讲  系统校准 (Calibration)

1. Introduction

2. Absolute orientation

3. Relative Orientation

4. Exterior Orientation

5. Interior Orientation

第十三讲  曲线与曲面 (Curves and Surfaces)

1. Basic Geometry

2. Representation of curve and surface

3. Surface Interpolation

4. Surface Approximation

5. Surface Segmentation

6. Surface Registration

第十四讲  动态视觉 (Dynamic Vision)

1. Change Detection

2.  3D structure from motion

3. Tracking

 

课程进度计划(Course Schedule):

 

课程考核要求(Course Examination Requirements)

作业(Homeworks):   10%

考试(Examination):   60%

课题研究报告(Research Report): 30%

 

参考文献(Course References)

1. Ramerh Jian, et al., Machine Vision, MIT Press and McGraw-Hill, Inc, 1995 (2003)

2. Dama H. Ballard, et al., Computer Vision, Prentice-Hall Inc., 1982

3. David Marr, Vision, W.H. Freeman and Company, 1982

4. Emanuele Trucco., Introductory Techniques for 3-D Computer Vision, Prentice-Hall Inc., 1998

5. 贾云得, 机器视觉，科学出版社，