| CLASSIFICATION OF MOBILE TERRESTRIAL LIDAR POINT CLOUDS IN URBAN AREA USING LOCAL FEATURES |
| Paper ID : 1339-SMPR |
| Authors: |
|
Mahdie Zaboli1, Heidar Rastiveis *2, Alireza Shams3, Benyamin Hosseiny1 1Department of Photogrammetry and Remote Sensing, School of Surveying and Geo-Spatial Engineering College of Engineering, University of Tehran, 2Department of Photogrammetry and Remote Sensing,
School of Surveying and Geo-Spatial Engineering
College of Engineering, University of Tehran, 3Postdoctoral Scholar, Advanced Highway Maintenance & Construction Technology (AHMCT) Research Center, University of California, |
| Abstract: |
| Automated analysis of three-dimensional (3D) point clouds has become a boon in Photogrammetry, Remote Sensing, Computer Vision, and Robotics. The aim of this paper is classifying an urban area point clouds, acquired by a Mobile Terrestrial Laser Scanning (MTLS) system, based on local geometrical and radiometric features. In this paper, the local features are firstly extracted for each point by observing their neighbor points. In this study, 14 features such as Linearity, planarity, mean of Intensity, etc. are implemented. The extracted features are then imported to a classification method to automatically label each points. Here, five powerful classification algorithms including k-Nearest Neighbours (k-NN), and Gaussian Naive Bayes (GNB), Support Vector Machine (SVM), Multilayer Perceptron (MLP) Neural Network, and Random Forest (RF) are tested. The points of the sample dataset were classified into eight semantic classes and the best overall accuracy was reported by RF algorithm equal to 87%. The results proved the reliability of the applied features for point clouds classification. |
| Keywords: |
| Mobile Terrestrial LiDAR, Point Clouds, Classification, Geometric Features |
| Status : Conditional Accept (Oral Presentation) |
