Please use this identifier to cite or link to this item: https://ah.nccu.edu.tw/handle/140.119/74452


Title: Road surface modeling from vehicle-borne point cloud by profile analysis
Authors: Wu, Chih-Wen;Chio, Shih-Hong
吳志文;邱式鴻
Contributors: 地政系
Keywords: Douglas-peucker algorithm;Feature constraints;LIDAR (light detection and ranging);Lidar systems;Mobile mapping systems;Point cloud;Position and orientation system(POS);Road surface model;Algorithms;Optical radar;Remote sensing;Three dimensional;Vehicles;Roads and streets
Date: 2012
Issue Date: 2015-04-10 15:25:05 (UTC+8)
Abstract: With the promotion of the accuracy of the position and orientation system (POS), mobile mapping system (MMS) is developed quickly. Especially, vehicle-borne LiDAR system combines the technology of the POS and LiDAR (light detection and ranging) to collect precise 3D point cloud of the detailed road corridor efficiently. However, the processing of the plenty of those point could is time-consuming. Therefore, those point cloud have to be preprocessed for post-processing more efficiently. Since the road surface is the most important information in the road corridor, this study will focus on the road surface modeling. First of all, the collected 3D point cloud will be organized and projected onto vertical profiles which are defined by the moving trajectory of vehicle-borne LiDAR system. Then, 3D projected points in each vertical profile will be further segmented into line segments by Douglas-Peucker algorithm. Next, the scene knowledge will be employed to extract line segment point on road surface. Finally, the boundaries of the road surface will be fitted by roubust estimation from the points on the road boundary extracted from road line segment points. From the results, it shows the feasibility of the proposed algorithm.
Relation: 33rd Asian Conference on Remote Sensing 2012, ACRS 2012,Volume 2, 1031-1040
Data Type: conference
Appears in Collections:[地政學系] 會議論文

Files in This Item:

File Description SizeFormat
index.html0KbHTML693View/Open


All items in 學術集成 are protected by copyright, with all rights reserved.


社群 sharing