A Kd-Tree-Based Outlier Detection Method for Airborne LiDAR Point Clouds

An outlier detection method is proposed based on the kd-tree for removing the outliers in the airborne LiDAR point clouds. In detailed, the kd-tree is employed to manage the airborne LiDAR data after the elimination of the obvious low and high outliers using the elevation histogram analysis, and for each point, the average of the distances between the central point and its A-neighborhood points are calculated. If the average distance is larger than an adaptively preset value, the point is regarded as an outlier. Eight datasets are utilized to test our method. Experiments show that our proposed method has many merits such as fewer input parameters, better performance and higher efficiency compared to typical method.