頁籤選單縮合
題 名 | 運用支持向量機建立土石流潛勢溪流判定模式=Using Support Vector Machine to Establish Models to Determine Potential Debris-Flow Creeks |
---|---|
作 者 | 沈哲緯; 羅文俊; 陳振宇; | 書刊名 | 水土保持學報 |
卷 期 | 43:2 2011.06[民100.06] |
頁 次 | 頁203-218 |
分類號 | 434.273 |
關鍵詞 | 3S 技術; 支持向量機; 土石流潛勢溪流; 3S technology; Support vector machine; Potential debris-flow creeks; |
語 文 | 中文(Chinese) |
中文摘要 | 本研究以農委會水土保持局98 年公開花蓮縣160 條土石流潛勢溪流為研究樣本,採用3S 技術萃取集水區地文因子,輔以水土保持局96~97 年土石流潛勢地區易致災因子調查計畫成果 與GIS 圖層,透過支持向量機(Support Vector Machine, SVM)建立花蓮縣土石流潛勢溪流判定規 則,預先透過分類模式判斷具有歷史災害之土石流溪流位置,並以土石流歷史災害目錄進行驗 證,其判定成效與土石流歷史災害目錄吻合程度高達九成以上,未來可供全臺土石流潛勢溪流 率定參考,協助主管機關進行災害防治、策略研擬及風險管理作業。 |
英文摘要 | This study used the 160 potential debris-flow creeks bulletined by Soil and Water Conservation Bureau(SWCB) as samples for analysis. Topographic factors were extracted using 3S technology,remote sensing (RS), geography information systems(GIS), and global positioning systems,(GPS). These were complemented by results from SWCB's hazard factors investigation program in the potential debris-flow areas as well as GIS layers. Through Support Vector Machine(SVM), a series of protocols to determine potential debris-flow creeks in Hualien County was established. The study determined the positions of the potential debris-flow creeks through classification model. The result was then verified by debris-flow hazard inventory. The match rate was over 90%. This may be expanded to recalculate the potential debris-flow creeks in Taiwan in the future. The results can assist the competent authorities to conduct disaster prevention, policy drafting, and risk management. |
本系統中英文摘要資訊取自各篇刊載內容。