DSpace Kochi University of Technology

Kochi University of Technology Academic Resource Repository >
A.学術情報資料別 >
SSMSパブリケーション >
Vol.08 >

Files in This Item:

File Description SizeFormat
sms12-7130.pdf1.34 MBAdobe PDFView/Open
Title: Applications of LiDAR measurement for road management
Authors: AKIYAMA, Shinpei
TAKAGI, Masataka
Issue Date: May-2012
Publisher: Society for Social Management Systems
Journal Title: Society for Social Management Systems Internet Journal
Abstract: Since LiDAR (Light Detection And Ranging) is a suitable equipment for archiving three-dimensional surface data of any objects. Moreover, aerial LiDAR is used for the topographical survey, urban planning or forest measurement. On the other hand, ground based LiDAR has a potential for other purposes, such as landslide monitoring or landcover change monitoring. This paper reports method of landslide and landcover monitoring using LiDAR for road management. Firstly, landslide monitoring technique using ground based LiDAR was developed. The amount of movements of a landslide should be precisely measured using LiDAR. In this study, Choja landslide in Japan was measured by LiDAR with a measurement accuracy of 6 mm. The result showed movement of landslide was detected in almost 1cm accuracy. An intersection point calculation of three surfaces was very effective for the accurate measurement. However, this technique can adapt for artificial object which include a plane. This method should be expand for natural objects such as natural slope or natural cliff. Secondary, landcover change is also detected by ground based LiDAR. A wide area landcover should be classified automatically using LiDAR. The landcover change can be extracted by converting grid model and comparing the elevation of the objects. Vegetated areas are changing seasonally and annually. The land cover change was detected in Choja. The results showed agricultural land could be extracted. Classified trees and Rice fields showed higher accuracy. LiDAR also acquires RGB color information. The classification will improve by using RGB color information in near future.
Type: Conference Paper
URI: http://hdl.handle.net/10173/1051
Appears in Collections:Vol.08

Please use this identifier to cite or link to this item: http://hdl.handle.net/10173/1051

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


Kochi University of Technology Library - Feedback