| 研究生: |
洪凱政 Hung, Kai-cheng |
|---|---|
| 論文名稱: |
應用多光譜影像多種特徵偵測崩塌地之研究 Landslide Detection Using Various Features from Multispectral Imagery |
| 指導教授: |
曾義星
Tseng, Yi-Hsing 徐百輝 Hsu, Pai-Hui |
| 學位類別: |
碩士 Master |
| 系所名稱: |
工學院 - 測量及空間資訊學系 Department of Geomatics |
| 論文出版年: | 2009 |
| 畢業學年度: | 97 |
| 語文別: | 中文 |
| 論文頁數: | 154 |
| 中文關鍵詞: | 影像分類 、物件導向 、知識庫 、變遷偵測 |
| 外文關鍵詞: | Knowledgebase, Change Detection, Object-Oriented, Image Classification |
| 相關次數: | 點閱:159 下載:13 |
| 分享至: |
| 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 |
台灣山坡地分佈極廣約佔全部面積百分之七十三,因山坡地地形陡峭、地質脆弱,加上台灣河川河短流急,平地之土地利用漸趨飽和,山坡地違規使用情形日益嚴重,每逢颱風、雨季來臨時,常有山崩、土石流等坡地災害,造成人民生命財產的嚴重危害。如何於災害發生後快速獲得災情資訊,並預測未來可能發生崩塌之危險區域,提早做好防災準備,一直是政府及相關單位所致力的目標之一。在各種災害監測技術中,遙測影像因為拍攝範圍廣、拍攝週期短,因此常被用來作為崩塌地判釋的主要技術之一。隨著遙測影像處理技術的發展,各種加強影像分類的方法也日新月異,其中引入物件導向模式以及建立知識庫分類的方法為目前研究的主要趨勢之一,如能有效利用這些新的分類方法,提升崩塌地判釋精確度,將可對坡地災情做更精確的估算與判斷。
本研究以艾利颱風為案例,利用颱風發生前後期的衛星影像以物件導向分類方式進行崩塌地偵測。在進行分類時,分別利用了影像像元之單一特徵及區塊物件之多種特徵。單一特徵通常為像元本身之光譜統計量,藉由監督式或非監督式等傳統分類方法,可決定該像元歸屬於何種地物類別,但其通常並未考量到相鄰像元間之相關性,以及地物形狀大小及紋理特徵等性質,當地物組成較複雜或地物光譜特徵性質相似時,容易有地物混淆的情況產生,因此本研究嘗試從區塊物件中挑選出與光譜、形狀、紋理等有關二十種特徵條件,並以物件導向分析方式,先建立特徵篩選機制以找出有利於分類之多種特徵條件,並建立知識庫,依循著合理的邏輯規則建立決策樹,以逐步分類的方式進行崩塌地判釋,以期達到提升判釋精度之目的。研究中亦採用其他常見之分類技術來偵測崩塌地,並進行分析比較,以找出最適合用於偵測崩塌地之方法與地物之特徵指標,以提升偵測崩塌地的精度與效率。
The main terrain type in Taiwan is slopeland which cover about 73% of total area. When the rainy season comes during typhoons, the landslides, mudslides and other disasters occurred frequently due to the steep terrain, rapid rivers, high geological vulnerability, and over development of slopelands. The landslope disasters always damage people as well as their property. How to quickly get the information of landslope disaster and predict the disaster in the future for preparedness has been the main study issue which the government and the relevant departments focus on. For disaster monitoring, satellite images are usually used to detect landslide locatons because such images contain larger range of areas and can re-capture the images of same place in a short period. Furthermore, with the development of the image processing technology, many novel classification methods, involoving the object-oriented-based and knowledge-based classification methods, are proposed. The accuracy of landslides detection would be expected be improved if these new classification methods could be applied effectively.
In this study, the object-oriented classification method is applied to detect the landslides from the satellite images which are captured separately before and after a typeoon event. Single feature extracted from a pixel and various features extract from objects are used to interpret the landslides. Single feature is usually derived from the spectral statistics of a single pixel and can be used in traditional supervised or unsupervised classification methods. However, the single feature does not consider the correlations between neighboring pixels. In addition, the sizes and the textures of objects do not be taken into account either. The result is that it can classify the objects hardly when the terrain is composed of many complicated or similar primitives. For this reason, various features derived from spectral, shape and texture of segmented objects are chosen in this thesis. The basic idea is based on object-oriented analysis. Firstly, the procedure of feature selecting is performed in order to find an appropriate condition to classify objects. Secondly, the knowledge database is established and be used to detect landslides according to the reasonable logic. This thesis also used many common techniques of change detection to recognize landslides, and the results were compared with the maximum likelihood method, object-oriented method in order to identify the most suitable classification method for satellite images and to improve the accuracy and efficiency of landslide detection.
江良印,紋理特徵應用於遙測影像判釋之理論研究,國立台灣大學農業工程研究所碩士論文,1998。
江美瑩等人,土石流、土石崩塌、高土沙含量河(溪水)之探討,國立台北大學自然資源與環境管理研究所報告,2004。
李三畏,台灣崩塌問題探討,地工技術,第7期,43-49頁,1984。
沈佩萱,多個母體的變異係數比較之有母數強韌法,國立中山大學碩士論文,2007。
林啟陽,地物導向自動化影像區格系統建立之研究,國立台灣大學碩士論文,1999。
林榮章,都會區多解像力遙測影像之紋理分析,國立中興大學土木工程學系研究所碩士論文,1999。
林家榮,潛在危險指標應用於屏東縣集水區分級分區之研究,屏東科技大學水土保持系碩士論文,2004。
林文賜、林昭遠、周文杰、黃碧慧,崩塌地自動萃取模式建立之研究,台灣地理資訊年會暨學術研討會,2005。
林中興,山坡穩定性評估之因子分析及地理資訊系統之應用,國立中央大學應用地質研究所碩士論文,1994。
林金炳、蔡光榮、侯峻棕、林昆賢、王嘉燁,"南橫公路甲仙-梅山路段潛在邊坡災害之調查分析",2001年,全國土地管理與開發學術研討會論文集。
林金樹,高光譜主軸轉換影像辨識土地利用型最適主軸數決定方法之研究,台灣林業科學 17(4):471-81,2002。
周明中,紋理輔助高解析度衛星影像分析應用於偵測入侵性植物分布之研究,國立中央大學碩士論文,2005。
洪皓仁,衛星影像分類方法之研究-以鳳山溪上游集水區為例,中興大學水土保持學研究所碩士論文,2000。
郭麟霂,寒帶沼地高光譜影像分類之研究,國立交通大學土木工程所,2000。
侯春帆,應用GIS及SPOT衛星影像於河川流域崩塌地淺勢評估研究_以陳有蘭溪為例,朝陽科技大學,2006。
孫彬修,線性複合模式應用於變遷偵測之研究,國立中央大學碩士論文,2004。
陳盈真,應用空間相干法在氣候模式之地物分類研究,國立中央大學碩士論文,2008。
陳嘉文,模糊邏輯在機械設計之應用,元智大學機械工程研究所碩士論文,1998。
陳振華、潘國樑,臺北市山坡地住宅區環境地質調查研究,工研院能源與礦業研究所報告,1985。
陳信雄、何智武、蔡光榮,烏山頭水庫集水區水土保持措施調查研究報告 中華水土保持學會,1988。
陳朝圳,南仁山森林生態系植生綠度之季節性變化,中華林學季刊,1999。
陳文福,以Landsat-TM及SPOT衛星影像監測高山地區土地利用變遷之研究,中華水土保持學報,1995。
高玉惠,小波轉換應用於影像自動判釋崩塌地分析,國立成功大學地球科學系碩士論文,2004。
唐德誠,灰度共現矩陣於多波段多極化SAR影像分類之研究,國防大學中正理工學院軍事工程研究所碩士論文,2002。
黃筱梅,SPOT衛星影像於裸露地變遷之偵測研究-以和社社區為例,國立台灣大學碩士論文,2001。
許中立,降雨滲透對邊坡穩定影響之研究,國立中興大學水土保持學研究所博士論文,1998。
莊雲翰,結合影像區塊及知識庫分類之研究—以IKONOS衛星影像為例,國立中央大學土木工程學系碩士論文,2002。
莊政斌,影像分割技術於高解析衛星影像分類之應用,國立中央大學碩士論文,2004。
張石角,山坡地潛在危險之預測及其在環境影響評估之應用,中華水土保持學報,1987。
楊智堯,類神經網路於邊坡破壞潛能分析之應用研究,國立成功大學土木工程學系碩士論文,1998。
楊龍士、周天穎,遙感探測理論與分析實務,逢甲大學地理資訊系統研究中心,2000。
楊永安,應用衛星影像進行坡地災害自動判釋與災因分析,台灣大學土木工程學系碩士論文,2007。
劉晃丞,應用高程差與福衛二號影像共同判釋新崩塌地,國立成功大學碩士論文,2007。
劉中河,類神經系統應用於白河水庫放流之研究,國立成功大學水利工程學系碩士論文,2004。
劉守恆,衛星影像於崩塌地自動分類組合之研究,國立成功大學碩士論文,2002。
廖浩鈞,使用QuickBird高解析度遙測影像以分類法判釋南二高邊坡保護工程之研究,國立成功大學碩士論文,2006。
廖軒吾,集集地震誘發之山崩,國立中央大學地球物理應用地質研究所碩士論文,2000。
謝漢欽、鄭祈全,福山地區SPOT多期影像植生綠度分析,林業試驗所研究報告季刊,1995。
黎瑋,紋理分析於遙測影像分類之研究,國立中央大學碩士論文,1997。
賴彥中,結合光達資料與數位空照影像重建三維建物模型,國立中央大學土木工程研究所碩士論文,2004。
賴君怡,整合空載光達資料與多光譜影像建立樹木模型-以成大校園為例,國立成功大學碩士論文,2008。
蕭國鑫,遙測技術應用於森林集水區內土砂災害之調查與監測-結合遙測與GIS應用於集水區災害監測,農委會94年度天然災害子計畫,2006。
蕭國鑫,劉治中,李惠容,遙測與GIS結合應用於德基水庫集水區土地利用/土地覆蓋,1994。
蕭國鑫、游明芳、劉進金、張志立,高解析影像應用於崩塌地判釋之研究, 台灣地理資訊年會暨學術研討會,2004。
蕭國鑫、尹承遠、劉進金、游明芳、王晉倫,SPOT影像與航照資料應用於崩塌地辨識之探討,航測及遙測學刊,第八卷,第四期,第29-42頁,2003。
鍾育櫻,921集集大地震前後降雨型崩塌地特徵之比較,國立台灣大學碩士論文,2005。
Baatz,M.,Benz,U.,Dehghani,S.,Heynen,M.,Holtje,A.,Hofmann,P.,Lingenfelder,I.,Mimler.,M.,Sohlbach,M.,and Weber,M,2004.eCognition Professional User Guide 4,Definiens Imagine GmbH,Munchen,Germany.
Baatz, M., and Schape, A,2000, “Multiresolution Segmentation – an optimization approach for high quality multi-scale image segmentation”, Angewandte Geographische Informationsverarbeitung XII, Beitrage zum AGIT-Symposium Salzburg 2000, Karlsruhe, Herbert Wichmann Verlag : 12-23.
Byrne GF,Crapper PF,MayoKK,1980.Monitoring land-cover change by principal component analys is of multitemporal Landsat data.Rem Sens Environ 10:175-84.
Chica-Olmo, M. and Arbarca-Hrnandez, F,2000. “Computing geostatistical image texture for remotely sensed data classification”, Computers and Geosciences, 26(4):373-383.
Choudhury, B.J,1994) Synergism of multispectral satellite observations for estimating regional land surface evaporation. Avenue of the Americas 49: 264-274.
DEFINIENS, eCognition object oriented image analysis ,User Guide 5.0.
Deutsch, C. V., and A. G. Journel ,1992, GSLIB: Geostatistical Software Library and User’s Guide, 340 pp., Oxford Univ. Press, New York.
Fernandez C. I.,Castillo T. F. D.,Hamdouni R. E. and Momtero J. C,1999 ‘Verification of Landslide Susceptibility Mapping: A Case Study.’ Earth Surface Processes and Landforms 24.
Green, E. P., Mumby, P. J., Edwards, A. J., Clark, C. D. and Ellis, A.C,1997, Estimating leaf area index of mangroves from satellite data. Aquat. Bot. 58:11-19.
Han , J. , Kamber ,M. “Data Mining: Concepts and Techniques” Morgan Kaufmann, 2000.
Haralick, R.M., K. Shaunmmugam, and I. Dinstein,1973.” Textural Features for Image Classification”, IEEE Trans. On Syst., Man, and Cybern., Vol. SMC-3, No.6, pp. 610-620.
Hay, G.J., K.O. Niemann, and G.F. Mclean,1996. “An Object-Specific Image- Texture Analysis of H-Resolution Forest Imagery”, Remote Sens. Environ., Vol. 55, pp.108-122.
Hughes, G.,1968. "On the mean accuracy of statistical pattern recognizers." IEEE Transactions on Information Theory, 14(1), 55-63.
Jakubauskas, M., K. P. Lulla, and P. W. Mausel,1990 Assessment of vegetation change in a fire-altered forest lanscape. Photogrammetry Engineering and Remote Sensing 56: 371-377.
Jensen, J. R.,1996. Introductory Digital Image Processing: A Remote Sensing Perspective. Englewood Cliffs, New Jersey: Prentice-Hall. 432pp. Jia, B., Z. Zhang, L. Ci, Y. Ren, B. Pan, and Z. Zhang,2004. Oasis land-use dynamics and its influence on the oasis environment in Xinjiang, China. Journal of Arid Environments 56: 11-26.
Kauth, R. J., and G. S. Thomas,1976, “The Tasseled Cap - A Graphic Description of the Spectral-Temporal Development of Agricultural Crops as Seen in Landdat, in Proceedings on the Symposium on Machine Processing of Remotely Sensed Data”, West Lafayette, Indiana, June 29 - July 1, (West Lafayette, Indiana: LARS, Purdue University), 41-51.
Lillesand, T. M. and R. W. Kiefer, 1994. Remote Sensing and Image Interpretation. 3nd Ed., John Wiley & Sons, New York.
Lyon, J.G. , D. Yuan,R.S. Lunetta, and C.D. Elvldge, “A Change Detection Experiment Using Vegetation Indices” , Photogrammetric Engineering & Remote Sensing,64(2):143-150 ,1998
Matsuyama,T.,1980,Structural Analysis of Natural Textures by Fourier Transformation. Computer Vision, Graphics and Image Processing, 12: 286-308.
Mockler R.J., 1992, Developing Knowledge-based Systems Using an Expert System Shell, New York, Macmillan Publishing Company.
Niemeyer, I. & M. J. Canty ,2001 : Knowledge-Based Interpretation of Satellite Data by Object-Based and Multi-Scale Image Analysis in the Context of Nuclear Verification. In: Proceedings of the IEEE International Geoscience and Remote Sensing Symposium, IGARSS'01, Sydney, Australia, 9-13 July 2001.
Oechel, W. C., and C. D. Reid ,1984Photosynthesis and biomass of chaparral shrubs along fire-induced age gradient in southern California. Bulletin de la Societe Chimique de France 131: 399-409.
Richards JA. 1984. Thematic mapping from multitemporal image data using the principal components transformation .Rem Sens Environ 16:35-46.
Price, J. C. and W. C. Bausch, 1995 Leaf area index estimation from visible and near-infrared reflectance data. Remote Sens. Environ. 52:55-65.
Rubec, C.D., and J. Thie., “Land use Monitoring with Landsat Digital Data in Southwestern Manitoba”, Proceedings of the fifth Canadian Symposium on Remote Sensing, Victoria, BC, 1987, pp. 136-150.
Schowengerdt, Robert. A. Remote Sensing: Models and Methods for Image Processing, 2nd edition,, Academic Press, 1997.
Specht, R. L.,1981 Primary productivity in Mediterranean-climate ecosystems regenerating after fire. p.257-267. In: Di Castri, D. W. F., and R. L. Specht (eds.) Mediterranean-type Shrublands. Elsevier, Amsterdam.
Stow, D. A., L. R. Tinney, and J. E. Estes, “Deriving Land Use/Land Cover Change Statistics form Landsat: A Study of Prime Agricultural Land”, Proceeding of the 14th International Symposium on Remote Sensing of Environment, pp. 1227-1237,1980.
Turker M. and E. Derenyi, 2000, GIS Assisted Change Detection Using Remote Sensing, Geocarto International, Vol.15,No.1,pp.49~54
Viedma, O., J. Melia, D. Segarra, and H. J. Garcia ,1997 Modelling rates of ecosystem recovery after fires by using Landsat TM data. Remote Sensing of Environment 61: 383-398.
Weismiller, R.A., S.J. Kristoof, D.K. Scholz, P.E. Anuta, and S.A. Momen, “Change Detection in Coastal Zone Environments”, Photogrammetric Engineering and Remote Sensing”, Vol.43, pp.1533-1539,1977.
Xiao,Q. and H. Raafat, “Remote Sensing Image Classification By A Gis Guided Spatial Analysis”, Geoscience and Remote Sensing Symposium, IGARSS '92. International on Pages: 1606 – 1608 ,1992
Zhao, L. , F.Y. Tan , W.Q. Quek , P. Chen ,and S.C. Liew, “Landuse Study of the Sentosa Island using SPOT Images” , IEEE International,(2):963-965 ,1997.