Feature Selection for Lidar Point-Cloud Classification based on Overlapping Regions

The global availability of dense point clouds provides the potential to better assess changes in our dynamic world, particularly environmental changes and natural hazards. A core step to make use of modern point clouds is to have a reliable classification and identify features of importance for a successful classification. Feature selection routines attempt to minimize the number of features by retaining as much information as possible about the classes. Our new approach attempts to achieve high classification results while being applicable to a wide variety of classifiers. 

We present an approach for classifier independent feature selection based on the overlap of features. The computative-extensive calculation of overlapping regions in multi-dimensional spaces is achieved by an optimized GPU-based Monte Carlo integration. This novel approach is compared against several feature selection routines and the selected features are tested with different classifiers. 

In our application experiments, we compare geometric, echo-based and full-waveform features of lidar point clouds to obtain the most useful sets of features for separating ground and vegetation points into their respective classes. Different scenarios of suburban and natural areas are studied to collect various insights for different classification tasks. In addition, we group features based on various attributes such as acquisition or computational cost and evaluate the benefits of these efforts in terms of a possible better classification result.