雷達影像於山區線型萃取與分析｜國立中央大學博碩士論文系統

簡易檢索 / 詳目顯示

回結果列表

研究生：	李明澤 Min-Tzer Lee
論文名稱：	雷達影像於山區線型萃取與分析 Using SAR image to extract lineament for investigating thetrend and distribution of the mountain range
指導教授：	吳宗達 Tzong-Dar Wu 陳錕山 Kun-Shan Chen
口試委員:
學位類別：	碩士 Master
系所名稱：	地球科學學院 - 太空科學研究所 Graduate Institute of Space Science
畢業學年度：	95
語文別：	中文
論文頁數：	73
中文關鍵詞：	斑駁雜訊、邊線偵測、多邊形估計、極化圖
外文關鍵詞：	polygonal approximation, edge detection, speckle, polarization plot
相關次數：	點閱：11 下載：0
分享至:	分享至facebook 分享至twitter

查詢本校圖書館目錄查詢臺灣博碩士論文知識加值系統勘誤回報

臺灣島因菲律賓海板塊與歐亞板塊不斷碰撞擠壓而形成，是屬於
山多平地少的地型，本研究之目的為:了解台灣山脈之走勢及其分
佈，研究中使用的資料為雷達影像(SAR image)並且對南台灣的山脈
做線型(lineament)偵測，萃取之山線再加以分析與統計其走勢之分佈。
研究中所使用的方法包含李氏濾波器(Refined Lee filter)來濾除
斑駁雜訊(speckle),邊線偵測(edge detection)方面則使用 Maximum
likelihood ratio edge detector 來強化輪廓線訊息，由檢測出來的線形
之中我們發現有斷線的情形及許多的非線形叢集(cluster)，在此使用
輪廓線追踨法(edge tracking)來消除斷線帶來的困擾，並使用標準差
(standard deviation)及數學型態學(Mathematical morphology)來剔除非
線性叢集，再應用鏈碼(chain code)及多邊形估計(polygonal
approximation)以直線的方式來逼近曲線，最後產生極化(polarization)
分佈圖描述山區線型走勢。
由研究中可以得知以下結論:從極化分佈圖來看，南台灣的主要
稜脈大致呈現北北東─南南西的走向，由此來推知山區受力的方向，
對局部小區域也做相同之檢視，可以發現到小區域山脈之受力方向則
不盡相同。
本研究主要的貢獻為提供主要山脈的分佈圖，分析大小尺度
(scale)的山區線形特徵，消除背景雜訊(clutter)及非線性的叢集，呈現
出主要山脈之走勢及其分佈，提供給地質學家(geologist)做為參考。

Due to the extrusion of Philippine Sea Plate and the Eurasian Plate,
the Taiwan Island was formed and the mountain areas are characterized
by less fiat ground. The purpose of this study is to understand the Taiwan.
SAR image was used for lineament investigating the trend and
distribution of the mountain range in the Southern Taiwan. Data collected
and computed to analyze the trend and distribution of the mountain range.
Research methods included applying Refined Lee filter to screen
speckle and using Maximum likelihood ratio edge detector for strengthen
information about the outline. From the results of investigation,
disconnection and cluster were found. Therefore, edge tracking was
applied to reduce the influence caused by disconnection. In addition,
Standard deviation and Mathematical morphology were utilized to
eliminate nonlinear cluster; chain code and polygonal approximation
were applied to approach to the curve and then polarization figure was
produced to illustrate the trend of mountain range.Conclusions drawn
from the research results are as follows: From polarization figure, the
distribution of Mountain range in the Southern Taiwan was a trend toward
NE- SW and the direction of loading could be referred according to this
figure. Checking on a small area of the mountain range, it appeared that
the directions of loading were different from one to the other.
The significance of this study is to provide with Figure of Mountain
distribution, analyze the characteristics of scales, eliminate clutter and
non-linear cluster, as well as present the trend and distribution of main
Mountain range. Those were references for geologists.

中文摘要.............................................………………………………….I
英文摘要......................................……………………………..…..……II
誌謝............................................……………………………………..…III
目錄..................................................…………………………………....IV
圖目錄...............................................…………………………………..VII
表目錄..............................................…………………………………...XII
第一章 緒論.......................……………………………….................….1
1 研究動機與目的…………………………………………………..1
2 研究流程與文獻回顧……………………………………………..2
3 本文架構…………………………………………………………..3
第二章 SAR 基本原理、斑駁雜訊抑制與邊線檢測器.......………….4
1 合成孔徑雷達系統................................……………………………4
1.1 頻率與穿透力..................................………………………..…4
2 系統參數.........................................………………………………...5
2.1 電磁波的極化.....................……………………………...........5
2.2 入射角................................………………………………........7
2.3 觀測幾何.............................……………………………….......7
3 地形參數......................……………………………………........…..9
3.1 地表目標物幾何特徵.........................………………………...9
3.2 地表粗糙度效應........………………………….................…...10
3.3 介電值（含水量）效應.......................……………………….10
3.4 紋理/組織分佈................…………………………..................11
IV
4 影像前處理………………………………………………………..11
4.1 輻射校正…………………………………………..…………11
4.2 幾何校正.....................................…………………..………...12
5 斑駁雜訊的成因..................................……………………...….…13
6 統計式濾波器..................................……………………………..14
6.1 LS 濾波器.................................………………………...…14
6.2 Refined Lee 濾波器..........................……………………....15
6.3 檢查濾波後之邊線是否偏移...................………………..…16
7 Maximum likelihood ratio 邊緣偵測器..............…………….....18
第三章 數學形態學與多邊形估計.....................………………..….21
1 數學形態學(Morphology).........................…………………..…21
1.1 膨脹(Dilation)與侵蝕(Erosion).................……………..….21
1.2 叢集標準差的計算...........................…………………..…..23
1.3 細線化.....................................………………………..……25
2 山區線型特徵之萃取.............................…………………..…….26
2.1 線追踨.....................................……………………..………26
2.2 累積能量之計算.............................………………..………27
2.2 使用標準差去除小凸枝.......................…………………....28
3 鏈碼與多邊形估計...............................……………………..….31
3.1 鏈碼的原理與應用...........................………………………31
3.2 點到直線距離...............................………………………....35
3.3 多邊形估計.................................………………………..…36
3.4 區域累積平方誤差(Local Integral Square Error) ...…..…..36
V
3.5 指北針的方位表示法與極化圖..................…………..……37
第四章 測試資料與成果展示...........................………………..…….39
1 實驗環璄與測試影像資料形式.......................…………..………39
1.1 實驗環璄....................................…………………..………..39
1.2 實驗資料及測區..............................……………..…………39
1.3 方法說明....................................………………..…………..41
2 山線分枝去除與資料向量化.........................………..…………..48
3 多邊形應用與最佳化...............................…………..……………49
4 山線走勢分佈極化圖及統計圖.......................……..……………52
5 實驗測區二......................................………………..…………….56
第五章 結論與展望.....................................……………..…………….66
1 結論..............................................………………….………………66
2 未來展望..........................................……………….………………66
參考文獻..........................................……………………………………68
附錄..............................................………………………………………70

                                

[1]Canny,J.F., "A computational approach to edge detection",IEEE Trans
Pattern Anal.Machine Intell,Vol.8,pp.679-698.,1986
[2]Lopes, A.,et al. "Adaptive speckle filters and scene homogeneity ",
IEEE TGARS, Vol. 28, No. 6, 1990.
[3]D.T. Kwan, et al. "Adaptive noise filtering for images with signal
dependent noise, " IEEE Pattern Analysis and Machine Intelligence,
Vol.7, No. 2, 1985.
[4]Frost, et al. "An adaptive filter for smoothing noisy radar images”,
IEEE Proceedings, vol.69,no. 1, 1981
[5]Gonzalez,w. " Digital Image Processing ", 2nd,pearson, pp.556-588 ,
2002.
[6]Lee,J.S."Digital image enhancement and noise filtering by use of local
statistics",IEEE Pattern Analysis and Machine Intelligence, Vol.2
No.2,1980.
[7]Lee,J.S."Refined filtering of image noise using local statistic",
Computer Vision,Graphics,and Image Processing Conference,
vol.15,380-389,1981.
[8]Lee,J.S.,I.Jorkevich,"Coastline Detection and Tracing in SAR images
",IEEE Trans.Geosci.Remote Sens.,Vol.28,No.4, July 1990.
[9]Otsu,N.,"A Threshold Selection Method from Gray-Level Histograms
",IEEE Transactions on Systems, Man, and Cybernetics, Vol.9,No.1,
1979,pp.62-66.
[10]Olivier G.and R.Philippe ,"On the bias of the likelihood ratio edge
detector for SAR images",IEEE TRANSACTION ON
68
GEOSCIENCE AND REMOTE SENSING ,VOL.38,NO. 3,2000.
[11]Rafael,C.,R.Gonzalez,and E.Woods,"Digital Image Processing", 2nd,
2002, pp. 519-560.
[12]William,K.P.,"Digital Image Processing",John Wiley & Sons, Inc,
2001, pp. 443-508.
[13]洪義凱,"合成孔徑雷達之地形線形特徵萃取",國立中央大學太
空科學研究所碩士論文,2005.
[14]張立奇,"應用數學形態學與合成孔徑雷達影像之線形特徵萃取",
國立海洋大學電機工程學系碩士論文,2006.
[15]鄧宜珍,"遙測影像處理與地貌辨識",國立中央大學資訊工程研
究所碩士論文,2003.
[16]鍾國亮, " 影像處理與電腦視覺 ", 2nd,pp.204-213,東華書局,2002
[17]徐啟崇, " 光學與合成孔徑雷達影像特徵分析於灘線萃取之研究",
國立中央大學資訊工程研究所碩士論文,1995.
69

簡易檢索 / 詳目顯示

相關論文