教學大綱與進度

課程基本資料：

學年期	課號	課程名稱	階段	學分	時數	修	教師	班級	人	撤	備註
110-2	301858	多維度影像處理	1	3.0	3	★	吳昭正	電機所 電資外生所	20	3	◎電機所和電資外生所合開

教學大綱與進度：

教師姓名	吳昭正
Email	ccwu@ntut.edu.tw
最後更新時間	2022-02-22 23:46:51
課程大綱	本課程屬於影像處理的進階課程，將聚焦於多光譜和高光譜影像分析與處理，並將涵蓋一般日常使用的彩色影像（多光譜影像的一種），以遙測影像與醫學影像作為主軸，分析各自不同的差異，以及所需的影像處理技術。 課程內容將涵蓋多光譜和高光譜影像的基礎、各種影像於分析時所面臨的挑戰以及所需的影像處理技術，包括線性分解、資料降維與升維、光譜與空間上的特徵萃取、機器與深度學習演算法。 選修本課程的學生需要具備基礎的信號處理、機率、線性代數和影像處理方面的背景。本課程將於學期初復習重要的基本概念，協助選修同學銜接本課程。 This is an advanced course on image processing. The main focus of this course is the analysis and processing of multispectral and hyperspectral images. It will cover the general color images as a type of multispectral image and be mainly based on remote sensing and medical images for analysis of their differences and corresponding image processing techniques. The scope of this course includesfundamentals of multispectral and hyperspectral imagery, challenges in image analysis, and corresponding image processing techniques, such as linear unmixing, dimensionality reduction, spectral and spatial feature extraction, and machine/deep learning algorithms. Studentsare expected to have a basic background in signal processing, probability, linear algebra, and image processing prior to taking this course. Important basic concepts will be reviewed at the beginning of the semester to help students lay the groundwork.
課程進度	Week 1(2/23). Introduction of multispectral and hyperspectral imagery Week 2(3/2). Sensors and data sets Week 3(3/9). Types of pixels and corresponding analysis Week 4(3/16). Unsupervised Analysis (I) Week 5(3/23). Unsupervised Analysis (II) Week 6(3/30). Supervised Analysis (I) Week 7(4/6). Supervised Analysis (II) Week 8(4/13). Dimensionality Reduction (I) Week 9(4/20). Dimensionality Reduction (II) Week 10(4/27). Band Selection (I) Week 11(5/4). Band Selection (II) Week 12(5/11). Endmember extraction (I) Week 13(5/18). Endmember extraction (II) Week 14(5/25). Abundance estimation (I) Week 15(6/1). Abundance estimation (II) Week 16(6/8). Classification (I) Week 17(6/15). Classification (II) Week 18(6/22). Presentations
評量方式與標準	Mini-Projects 60% Oral Presentation 20% Class Participation 20%
使用教材、參考書目或其他	【遵守智慧財產權觀念，請使用正版教科書，不得使用非法影印教科書】 使用外文原文書：是 Hyperspectral Data Processing: Algorithm Design and Analysis, 1st Edition, Chein-I Chang, Wiley, 2013.
課程諮詢管道
備註	If the class needs to be online due to COVID-19, the link and related codes will be delivered to your university's email. Please contact me by ccwu@ntut.edu.tw if you have any further question.