Please use this identifier to cite or link to this item:
標題: Application of a High-Resolution Digital Surface Model Produced by an Unmanned Aerial Vehicle in Canopy Gap Interpretation and Slope Stability Assessment
無人載具產製之高精度數值地表模型應用於 樹冠孔隙自動判釋及邊坡土體穩定評估
作者: Wei-Kai Huang
I Ting
Ching-Fang Lee
關鍵字: unmanned aerial vehicle;digital surface model;semiautomatic interpretation;canopy gap;slope stability;UAV;DSM;樹冠孔隙;半自動判釋;邊坡土體穩定
Project: 中華水土保持, Volume 48, Issue 4, Page(s) 175-187.
This study establishes a relationship between the density of canopy gaps and slope stability by using a high-resolution digital surface model (DSM) made from images recorded by an unmanned aerial vehicle (UAV) to semiautomatically estimate canopy gap density. Estimated gap density can be employed to identify potentially unstable slopes. The forested slopes adjacent to Xinzhuang Village in southern Taiwan were used as a case study. The estimated canopy gap density is consistent with that of the field investigation. Where the canopy gap density or slope gradient is low, the slope surface is generally undisturbed. Where canopy gaps are dense, tension cracks and scars are often present, bedrock orientation is more erratic, and gullies are more densely spaced on the slope surface. At the Xinzhuang village, a strongly negative relationship was identified between canopy irregularities detected in the DSM and slope instability. Canopy gap density detected with high-resolution DSM may therefore be a useful reference for preliminary slope hazard mapping.

本研究目的旨在建立樹冠孔隙密度與邊坡土體穩定關係,以無人載具拍攝影像所產製之 DSM,以 半自動判釋萃取出樹冠孔隙並分析其密集程度,瞭解邊坡潛在之不穩定區位。本研究以高雄市杉林區新庄里 境內之獅子山周緣坡地為案例,根據樹冠孔隙密度分析成果顯示,樹冠孔隙密度較低之區域,植生林相多為 平整一致或為平地地形;而樹冠孔隙較密集之區域,經比對現場狀況,發現林相多為倒塌紊亂,並存有張力 裂縫及崩崖,露頭量測位態也與區域位態差異甚大;另蝕溝流經之區域,該樹冠孔隙密度也相對較高。顯示 坡體經擾動後與樹冠孔隙密集程度呈負相關,該關聯性可初步提供重點調查及坡體破壞熱點之參考。
Appears in Collections:第48卷 第04期

Files in This Item:
File Description SizeFormat
97350-3.pdf14.1 MBAdobe PDFView/Open
Show full item record

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.