教學大綱表 Syllabus
請遵守智慧財產權,勿使用非法影印教科書,避免觸法。
課程名稱
Course Title
(中文) 電腦視覺
(英文) Computer Vision
開課單位
Departments
資訊工程研究所
課程代碼
Course No.
I5870
授課教師
Instructor
謝禎冏
學分數
Credit
3.0 必/選修
core required/optional
選修 開課年級
Level
研究所
先修科目或先備能力Course Pre-requisites:資料結構、英文聽與說
課程概述與目標 Course Overview and Goals:Course overview and objectives: Computer vision is the science of how to make machines "see." In other words, it uses cameras and computers instead of human eyes to detect, identify, track and measure targets as the basis for intelligent system decision-making. The lecture content includes the basic concepts, theoretical basis, and possible applications of computer vision. This course will also guide students to use the OpenCV computer vision library to implement and simulate several computer vision topics. This course is taught in EMI.
教科書
Textbook
自編教材
參考教材
Reference
1. Computer Vision by Linda G. Shapiro and George C. Stockman
2. Learning OpenCV: Computer Vision with the OpenCV Library by Gary Bradsk
課程大綱
Syllabus
學生學習目標
Learning Objectives
單元學習活動
Learning Activities
學習成效評量
Evaluation
備註
Notes

Week
單元主題
Unit topic
內容綱要
Content summary
1 Introduction Computer Vision and Applications Understand why you need to learn computer vision.
  • 講授
  •  
    2 Camera Model and Image Representation 1. 2D image from projection of a 3D scene
    2. Imaging formation
    3. Pin-hole camera
    4. Video cameras
    5. Lab: save images by OpenCV
    Fundamental image processing
  • 實作
  • 講授
  • 上機實習
  •  
    3 Filtering and Enhancing Images 1. Gray level transformations
    2. Histogram processing
    3. Enhancement using arithmetic operations
    4. Spatial filtering
    5. Lab: image enhancement
    Learn how to call OpenCV for image preprocessing
  • 實作
  • 講授
  • 上機實習
  •  
    4 Binary Image Processing 1. Pixels and neighborhood
    2. Applying masks to images
    3. Counting the object in an image
    4. Connected components labeling
    5. Binary image morphology
    6. Region properties
    7. Region adjacency graphs
    8. Thresholding gray-scale images
    9. Lab: Connected Component
    Feature extraction
  • 討論
  • 實作
  • 講授
  •  
    5 Case study - OCR 1. Handwritten OCR systems
    2. CIL - Greek Handwritten Character Database
    3. Proposed OCR Methodology
    4. Experimental Results
    5. Experiments on Historical Documents
    6. Lab: OCR
    Learn how to write code for OCR
  • 討論
  • 實作
  • 講授
  •  
    6 Image Segmentation 1. Clustering
    2. K-means
    3. Region Growing
    4. Lab: Segmentation
    Ideally, partition an image into regions corresponding to real world objects
  • 討論
  • 實作
  • 講授
  • 作業 Assignment
  •  
    7 Case study - Multi-Focus Images Fusion 1. Region-based image fusion methods
    2. Multifocus images fusion by average
    3. Intermediate fused image is segmented
    4. Source images are segmented according to the segmenting result
    5. Segmented regions of the source images are fused
    6. Lab: Multi-Focus Images Fusion
    Multifocus image fusion using region segmentation and spatial frequency
  • 個案研究
  • 討論
  • 講授
  • 作業 Assignment
  •  
    8 Case study –High Dynamic Range Image 1. What is HDR?
    2. Display HDR image by different exposure level
    3. Tone mapping
    4. Lab: HDRI Fusion
    Learn how to do a high dynamic range image.
  • 個案研究
  • 講授
  •  
    9 Mid-Term Exam Papaer Exam. Pass the exam.
  • 期中考 Midterm
  •  
    10 Case study - Motion from 2D Image Sequences 1. 2-D motion vs. optical flow
    2. Motion representation
    3. Motion estimation criterion
    4. Optimization methods
    5. Gradient descent methods
    6. Lab: Optical Flow
    Motion estimation
  • 討論
  • 講授
  •  
    11 Intelligent Video Surveillance 1. Industrial need and trend
    2. Aplication scenario and technology development
    3. ITRI developed technologies and industrial applications
    Understand industrial trend and requirement
  • 討論
  • 講授
  •  
    12 Case study - Behavior Analysis 1. Motion segmentation
    2. Morphological operation
    3. Object classification
    4. Tracking
    5. Lab: Tracking
    Learn how to do video surveillance
  • 實作
  • 講授
  •  
    13 Object Matching - SIFT 1. Scale Space and Difference of Gaussian
    2. Key point Localization
    3. Orientation Assignment
    4. Descriptor Building
    5. Application
    6. Lab: Object Recognition
    Learn how to do object matching using SIFT
  • 討論
  • 實作
  • 講授
  •  
    14 Pattern Recognition Concept 1. Common model for classification
    2. Precision vs. recall
    3. Feature vector representation
    4. Implementing the classifier
    5. Structural techniques
    6. The confusion matrix
    7. Decision trees
    8. Matching by relations
    9. Lab: Neural Network
    Understand techniques of pattern recognition
  • 討論
  • 講授
  •  
    15 Face Detection and Recognition 1. Face feature
    2. Integral image
    3. Ada-boost
    4. Face detection
    5. Lab: Face Detection
    Learn how to write programs to detect human faces.
  • 實作
  • 講授
  •  
    16 Gesture Recognition 1. Image processing methods
    2. Image pyramid
    3. human face detection
    4. digital focusing
    5. Adaptive skin color detection
    6. motion history image
    7. Lab: Hand Gesture Recognition
    Learn how to detect hand gesture
  • 實作
  • 講授
  •  
    17 Special Topic Select a research topic Final project-Search published papers and design a workable computer vision system
  • 設計研究
  • 討論
  • 心得發表
  •  
    18 Term project Paper exam. or give a final report (oral and a ppt file) Pass
  • 期末考 Final Exam
  •  

    教學要點概述 Overview of Teaching Points:
    教材編選 Teaching Materials: ■ 自編教材 Handout by Instructor □ 教科書作者提供 Textbook
    評量方法 Evaluation: 期末考 Final Exam:30%   期中考 Midterm:30%   問答 Interrogation:5%   上機測驗 Examination on Computer Operation:35%  
    教學資源 Teaching Resources: ■ 教材電子檔 Soft Copy of the Handout or the Textbook ■ 課程網站 Website
    課程網站:網路大學
    扣考規定 The rule of being able to take the final exam of the course:http://eboard.ttu.edu.tw/ttuwebpost/showcontent-news.php?id=504