รหัสดีโอไอ 10.14457/TU.the.2019.760
Title Automated recognition of road structures and road facilities from mobile laser scanning point clouds: a case study in Bangkok Expressway system
Creator Tran Thanh Ha
Contributor Taweep Chaisomphob, Advisor
Publisher Thammasat University
Publication Year 2019
Keyword Point clouds ,Mobile laser scanning ,Road structures ,Road facilities ,Recognition ,Classification
Abstract A road network is used for the daily activities of citizens including transport services, goods, and people. Expressway/highway systems are an integral part of the road network, which is designed to handle traffic congestion, snarl–up and traffic collision inside the city and also reduce the travel time from city to city. Recently, remote sensing in form of aerial and satellite images and laser scanning was emerging advanced technologies, allowing capture the large–scale three–dimensional (3D) topographic data of the road network accurately and quickly. Ones of which, laser scanning integrated into a vehicle, called mobile laser scanning (MLS) was to be a great unit for capturing 3D road information at a large scale effectively and precisely (8,000 points/m2 with absolute accuracy about 5 mm). Using point cloud data collected from MLS, all road information can be obtained. The collected expressway information data consisted of expressway facilities (including lighting poles, noise barriers, traffic signs, overhead signs, telecommunication stations, emergency phones, traffic safety equipment) and expressway surfaces features (including manholes, road slope, lane’s width, lane’s number, road markings, road curbs and road cracks status). Each road information can be extracted individually for maintenance and asset management. However, manual extraction of road information is time-consuming due to the bulk of point clouds. Thus, it is essential to propose a method that can extract road data in an automated and effective way by inheriting the advantages of MLS.This study presented a voxel-based region growing method for automatic road surface extraction from mobile laser scanning point clouds in the expressway environment. The proposed method contained three major steps: constructing a voxel model; extracting the road surface points by employing the voxel based segmentation algorithm; refining the road boundary using the curb–based segmentation algorithm. From the extracted road surface, the framework for the estimation of road roughness was proposed. A good agreement was achieved when compared the results of road roughness map to the visual image, thereby indicating the feasible and effective of the proposed algorithm.The automated localization and classification of road facilities are also proposed in this study. First, the ground points are eliminated, and the non–ground points are merged into clusters. Second, the pole-like objects are extracted using horizontal crosssection analysis and minimum vertical height criteria. Finally, a set of knowledge-based rules, which comprises height features and geometric shape, is constructed to classify the detected road poles into different types of road facilities. Two test sites of point clouds in an expressway environment, which are located in Bangkok, Thailand, are used to assess the proposed method. The first test site is a part of Bang Na– Bang Phli–Bang Pakong (BangNa) expressway which is the longest elevated highway in Thailand with the total surveyed length of 5.1 km. The second test site was conducted at cable–stayed Rama IX Bridge which is a part (2.1 km) of the most congested national expressway. Both expressways were operated by the Expressway Authority of Thailand (EXAT). The results show that the proposed method could be a promising alternative for the extraction of road surface and road facilities with acceptable accuracy.
ดิจิตอลไฟล์ Digital File #1

บรรณานุกรม

Tran Thanh Ha และผู้แต่งคนอื่นๆ. (2019) Automated recognition of road structures and road facilities from mobile laser scanning point clouds: a case study in Bangkok Expressway system. Thammasat University:ม.ป.ท.
Tran Thanh Ha และผู้แต่งคนอื่นๆ. 2019. Automated recognition of road structures and road facilities from mobile laser scanning point clouds: a case study in Bangkok Expressway system. ม.ป.ท.:Thammasat University;
Tran Thanh Ha และผู้แต่งคนอื่นๆ. Automated recognition of road structures and road facilities from mobile laser scanning point clouds: a case study in Bangkok Expressway system. ม.ป.ท.:Thammasat University, 2019. Print.