作者: Francesco Pirotti , Paolo Tarolli
DOI: 10.1002/HYP.7582
关键词: Geology 、 Point (geometry) 、 Curvature 、 Smoothing 、 Standard deviation 、 Remote sensing 、 Lidar 、 Range (statistics) 、 Communication channel 、 Landform
摘要: This study uses landform curvature as an approach for channel network extraction. We considered a area located in the eastern Italian Alps where high-quality set of LiDAR data was available and heads related were mapped field. In analysis, we derived 1-m DTMs from different ground point densities, used smoothing factors landscape calculation order to test suitability density maps recognition network. methodology is based on threshold values calculated multiples (1–3 times) standard deviation curvature. Our analyses suggested that (i) window size calculations has be function features detected, (ii) coarse could useful finer one main (iii) rougher are not optimal they do explore sufficient range at which occur, while smoother overcome this problem more appropriate extraction surveyed channels. Copyright © 2010 John Wiley & Sons, Ltd.