Feasibility of Multispectral Airborne Laser Scanning Data for Road Mapping

Authors:
Open Access
Publisher website
Google Scholar
Abstract
Multispectral airborne laser scanning (ALS) data have recently become available. The objective of this letter is to study the feasibility of these data for road mapping-for road detection and road surface classification. The results are compared with the results of traditional aerial ortho images using object-based image analysis and Random Forest classification. The results demonstrate that the multispectral ALS data are feasible for automatic road detection and a significant improvement compared to the use of optical aerial imagery is obtained. In a test using ALS data, 80.5% points representing roads were classified correctly. When aerial images were used, the percentage decreased to 71.6%.
Funding Information
  • Research Council of Finland (295047 (“Integration of large multisource point cloud image datasets for adaptive map updating”), 272195 (“Centre of Excellence in Laser Scanning Research”))

This publication has 21 references indexed in Scilit: