Abstract
Object-based point cloud analysis (OBPA) is useful for information extraction from airborne LiDAR point clouds. An object-based classification method is proposed for classifying the airborne LiDAR point clouds in urban areas herein. In the process of classification, the surface growing algorithm is employed to make clustering of the point clouds without outliers, thirteen features of the geometry, radiometry, topology and echo characteristics are calculated, a support vector machine (SVM) is utilized to classify the segments, and connected component analysis for 3D point clouds is proposed to optimize the original classification results. Three datasets with different point densities and complexities are employed to test our method. Experiments suggest that the proposed method is capable of making a classification of the urban point clouds with the overall classification accuracy larger than 92.34% and the Kappa coefficient larger than 0.8638, and the classification accuracy is promoted with the increasing of the point density, which is meaningful for various types of applications.
Highlights
Recent years have seen the emergence of airborne LiDAR (Light Detection And Ranging), termed as airborne laser scanning (ALS), as a leading technology for the extraction of information about physical surfaces [1]
Point cloud classification is one of the hot topics in the field of ALS data processing, because classification is a precondition of many applications such as building reconstruction or vegetation modeling or flood modeling [8].Point cloud classification implies the separation of points into vegetation, building or ground classes, and each of these classes implies the knowledge of its nature
After reviewing the existing methods, we found that the segment-based and simultaneous classification is becoming more and more popular, which should attribute to the point cloud segmentation
Summary
Recent years have seen the emergence of airborne LiDAR (Light Detection And Ranging), termed as airborne laser scanning (ALS), as a leading technology for the extraction of information about physical surfaces [1]. Compared with other techniques such as interferometric SAR and photogrammetry, ALS has the advantage of acquiring dense, discrete, detailed and accurate 3D point coverage over both the objects and ground surfaces directly. The detailed depiction of objects and ground surfaces by the dense point cloud may indicate a directness or simplicity in which objects and related information can be retrieved from the data. Similar to the existing data sources such as aerial or satellite optical imagery, extensive manual post-processing is still required to extract accurate terrain or semantic information from the LiDAR point clouds [7], and underlining the necessity for research of the automatic methods in this area. Lots of point cloud classification methods have been proposed, and there are different classification systems for the existing methods based on different criteria
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
