Classification Method for Object Feature Extraction Based on Laser Scanning Data

Mobile laser scanning has many advantages compared to traditional survey technology. These characteristics make it possible to rapidly acquire large-area high-precision 3D spatial data for reconstruction of 3D model. In this paper, we focus on the classification and recognition of mobile laser scanning data. We present a method based on object feature extraction. The main workflow of this method is made up by extraction of geometric feature from scanning data,foundation of objects feature knowledge database, extraction of main feature by PCA (Principal Component Analysis) to match geometric features,multi-level classification according to object’s principal feature at last. This method has been applied to data points obtained by mobile laser scanning system; the experiment results show that the proposed classification method is promising.