![]() |
|
以作者查詢圖書館館藏 、以作者查詢臺灣博碩士 、以作者查詢全國書目 、勘誤回報 、線上人數:29 、訪客IP:3.135.188.194
姓名 李光哲(Kuang-Che Lee) 查詢紙本館藏 畢業系所 太空科學研究所 論文名稱 利用遙測影像自動萃取校正點
(Automatic Tie Point Extraction for Remote Sensing Image)相關論文 檔案 [Endnote RIS 格式]
[Bibtex 格式]
[相關文章]
[文章引用]
[完整記錄]
[館藏目錄]
[檢視]
[下載]
- 本電子論文使用權限為同意立即開放。
- 已達開放權限電子全文僅授權使用者為學術研究之目的,進行個人非營利性質之檢索、閱讀、列印。
- 請遵守中華民國著作權法之相關規定,切勿任意重製、散佈、改作、轉貼、播送,以免觸法。
摘要(中) 近年來,遙測影像的應用範圍愈來愈廣泛,在製作地圖方面經常會使用到,甚至在工程上也有使用,但是,在使用遙測影像之前還需要經過一些處理,其中最重要的就是幾何校正這方面,一般來說,影像校正是使用載具上所記錄的飛行軌道資訊來進行,但是若無軌道資訊時,就會取一張level-4影像,在兩張影像相同地點上選取校正點。而不同載具所使用的光譜波段不盡相同,不同光譜的反射量值對相同物體也不一樣,不過在現行的自動選取校正點方法中,通常是使用相關係數來選出同樣的地區,再從這裡找出特徵點,可是這方法在兩張不同載具所拍攝的影像上就無法使用,原因在於相關係數是使用反射量值來做選取,光譜不同的話,找出來的區域就很可能不符,所以這份研究是希望找出一個方法,讓不同來源的影像可以找出相同位置的校正點,藉此減少人工找點的時間。
此論文使用 ”完全限制最小平方差” (Fully Constrained Least Square) 分類法來萃取出地面上的資訊,如道路和湖泊,再從這些資訊中,找出特徵點做為校正點,如十字路口和湖泊幾何中心。
摘要(英) Remote sensing image become more and more popular in near years. It also is used in engineering and making map. But the image can not use directly when we get the raw image data (level-1 image). The main method to solve this problem is using satellite’s flight information to do geometry correction. If we do not have this information, we can take one image which was corrected (level-4 image) to decide the tie points. Generally speaking, the tie points automatic extracting is using correlation in two different level images to find same feature being tie points. Correlation method is using reflectivity to select same area. It can not use correlation to decide the same position, because the reflectance of one material would different in each band. And the wave bands which different satellites used are not always being same bands. In this research, we want to find a method to decide the tie points in different sensor images.
We use the Fully Constrained Least Square (FCLS) classification to extract the information on ground. Our opinion is using the information on the ground to find the same feature, like roadd and lakes. Then detecting cross road and barycenter of lake to be tie points. We hope this method can help operator find the tie points easier.
關鍵字(中) ★ 校正點
★ 遙測影像
★ 完全限制最小平方差關鍵字(英) ★ remote sensing image
★ FCLS
★ tie points論文目次 ABSTRACT (Chinese) …...…………………………………………..…i
ABSTRACT (English) …...…………………………………………..…ii
CONTENTS ……………………………………………...…………….iii
CONTENTS OF FIGURES..………………...……………………...….v
CONTENTS OF TABLE ..……………………….…………………..viii
CHAPTER
1. Introduction……………………………………………...………...…1
1.1 Motivation and Flow Chart…...………………………………...…1
1.2 Background………………………………………………………..3
1.3 Thesis Organization……………………………………………….4
2. Data Extraction…………………………………………….………..5
2.1 Image Fusion…………………..…………………….…………….5
2.2 Fully Constrained Least Square..………………………………….6
2.2.1 Linear Mixture Model………………………...……………....6
2.2.2 Least Square…………………………………………………..7
2.2.3 Sum-to-one Constrained Least Square………………………..8
2.2.4 Nonnegative Constrained Least Square……………………….8
iv
2.2.5 Fully Constrained Least Square……………………………….9
2.3 Gaussian-pyramid………………...………………………………..9
2.4 Mathematical Morphology………………………………............10
3. Tie Points Extraction……………………………………………….12
3.1 Decide Tie Points by Cross Road………………………………...12
3.1.1 Thinning Method…………………………………..……...…13
3.1.2 Cross Road Detected…………………………….….……….15
3.2 Decide Tie Points by Lakes……………………………….….…..18
3.2.1 Region Growing and Morphological Closing……………….19
3.2.2 Find Tie Points by Barycenters………………………………21
4. Experiment Results and Discussions……………………………..23
4.1 Satellite Introduction………………………………….................23
4.2 Experiment Result of Cross Road Detect…………...……………28
4.3 Experiment Result of Lake’s Barycenter Detect…………………46
5. Conclusions and Future Work…………………………………...60
References……………………………………………………………61
參考文獻 [1] Daniel C. Heinz, Chein-I Chang, “Fully Constrained Least Squares Linear Spectral Mixture Analysis Method for Material Quantification in Hyperspectral Imagery”, vol.39, No.3, p.529-545, March 2001, IEEE transaction on geosciences and remote sensing
[2] Chun-Ming Tsai, Zong-Mu Yeh, “Contrast Enhancement by Automatic and Parameter-Free Piecewise Linear Transformation for Color Images”, vol. 54, issue 2, p.213-219, 2008, IEEE transactions on consumer electronics
[3] 潘彥男, “利用電腦自動化對數值高程模作線形偵測”, 任玄, 2007 六月,國立中央大學碩士論文
[4] Louisa Lam, Seong-Whan Lee, Ching Y. Suen, “Thinning Methodologies-A Comprehensive Survey”, vol.14, No. 9, September 1992, IEEE transactions on pattern analysis and machine intelligence
[5] Gonzalez, Rafael C.,Woods, Richard E., Eddins, Steven L., “Digital Image Processing using Matlab”, September 2005, Tung Hua
[6] Lam, L., Seong-Whan Lee, and Ching Y. Suen, "Thinning Methodologies-A Comprehensive Survey," IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol 14, No. 9, p.879,September 1992
[7] Jiuxiang Hu, Anshuman Razdan, John C. Femiani, Ming Cui, and Peter Wonka, “Road Network Extraction and Intersection Detection From Aerial Images by Tracking Road Footprints”, vol. 45, No. 12, p.4144-4157, December 2007, IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
指導教授 任玄(Hsuan Ren) 審核日期 2010-7-16 推文 plurk
funp
live
udn
HD
myshare
netvibes
friend
youpush
delicious
baidu
網路書籤 Google bookmarks
del.icio.us
hemidemi
myshare