Eigen-feature analysis of weighted covariance matrices for LiDAR point cloud classification |
| |
| Institution: | 1. Department of Geomatics, National Cheng Kung University, Taiwan;2. Department of Electrical and Computer Engineering, Old Dominion University, USA;1. State Key Laboratory of Remote Sensing Science, Beijing Key Laboratory for Remote Sensing of Environment and Digital Cities & Faculty of Geography, Beijing Normal University, China;2. Department of Geographical Sciences, University of Maryland, College Park, MD, USA;3. Photogrammetry and Remote sensing, Technische Universitaet Muenchen, Munich, Germany;1. Université Paris-Est, IGN Recherche, SRIG, MATIS, 73 avenue de Paris, 94160 Saint Mandé, France;2. Centre de Morphologie Mathématique (CMM), 35 rue Saint Honoré, 77305 Fontainebleau, France;1. Faculty of Geosciences and Environmental Engineering, Southwest Jiaotong University, Gaoxin West District, Chengdu, China;2. Collaborative Innovation Center for Geospatial Technology, 129 Luoyu Road, Wuhan, China;3. Department of Land Surveying and Geo-Informatics, The Polytechnic University of Hong Kong, Hum Hom, Kowloon, Hong Kong |
| |
| Abstract: | The features used in the separation of different objects are important for successful point cloud classification. Eigen-features from a covariance matrix of a point set with the sample mean are commonly used geometric features that can describe the local geometric characteristics of a point cloud and indicate whether the local geometry is linear, planar, or spherical. However, eigen-features calculated by the principal component analysis of a covariance matrix are sensitive to LiDAR data with inherent noise and incomplete shapes because of the non-robust statistical analysis. To obtain reliable eigen-features from LiDAR data and to improve classification accuracy, we introduce a method of analyzing local geometric characteristics of a point cloud by using a weighted covariance matrix with a geometric median. Each point is assigned a weight to represent its spatial contribution in the weighted principal component analysis and to estimate the geometric median which can be regarded as a localized center of a shape. In the experiments, qualitative and quantitative analyses on airborne LiDAR data and simulated point clouds show a clear improvement of the proposed method compared with the standard eigen-features. The classification accuracy is improved by 1.6–4.5% using a supervised classifier. |
| |
| Keywords: | Point cloud classification Weighted covariance matrix Eigen-feature |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
