Fast Filtering of LiDAR Point Cloud in Urban Areas Based on Scan Line Segmentation and GPU Acceleration

The fast filtering of massive point cloud data from light detection and ranging (LiDAR) systems is important for many applications, such as the automatic extraction of digital elevation models in urban areas. We propose a simple scan-line-based algorithm that detects local lowest points first and treats them as the seeds to grow into ground segments by using slope and elevation. The scan line segmentation algorithm can be naturally accelerated by parallel computing due to the independent processing of each line. Furthermore, modern graphics processing units (GPUs) can be used to speed up the parallel process significantly. Using a strip of a LiDAR point cloud, with up to 48 million points, we test the algorithm in terms of both error rate and time performance. The tests show that the method can produce satisfactory results in less than 0.6 s of processing time using the GPU acceleration.

[1]  K. Kraus,et al.  Determination of terrain models in wooded areas with airborne laser scanner data , 1998 .

[2]  G. Vosselman.  SLOPE BASED FILTERING OF LASER ALTIMETRY DATA , 2000 .

[3]  P. Axelsson.  DEM Generation from Laser Scanner Data Using Adaptive TIN Models , 2000 .

[4]  I. Dowman,et al.  TERRAIN SURFACE RECONSTRUCTION BY THE USE OF TETRAHEDRON MODEL WITH THE MDL CRITERION , 2002 .

[5]  Chengcui Zhang,et al.  A progressive morphological filter for removing nonground measurements from airborne LIDAR data , 2003, IEEE Trans. Geosci. Remote. Sens..

[6]  Nicolas H. Younan,et al.  DTM extraction of lidar returns via adaptive processing , 2003, IEEE Trans. Geosci. Remote. Sens..

[7]  George Vosselman,et al.  Experimental comparison of filter algorithms for bare-Earth extraction from airborne laser scanning point clouds , 2004 .

[8]  J. Shan,et al.  Urban DEM generation from raw lidar data: A labeling algorithm and its performance , 2005 .

[9]  George Vosselman,et al.  Segmentation of point clouds using smoothness constraints , 2006 .

[10]  Le Wang,et al.  A multi-resolution approach for filtering LiDAR altimetry data , 2006 .

[11]  Gustav Tolt,et al.  Segmentation and classification of airborne laser scanner data for ground and building detection , 2006, SPIE Defense + Commercial Sensing.

[12]  T. Rabbani,et al.  SEGMENTATION OF POINT CLOUDS USING SMOOTHNESS CONSTRAINT , 2006 .

[13]  P. Gong,et al.  Filtering airborne laser scanning data with morphological methods , 2007 .

[14]  Kiyun Yu,et al.  An Approach for Segmentation of Airborne Laser Point Clouds Utilizing Scan-Line Characteristics , 2007 .

[15]  Dah-Jye Lee,et al.  Real-Time Optical Flow Calculations on FPGA and GPU Architectures: A Comparison Study , 2008, 2008 16th International Symposium on Field-Programmable Custom Computing Machines.

[16]  K. Clint Slatton,et al.  Fast Real-Time LIDAR Processing on FPGAs , 2008, ERSA.

[17]  Kiyun Yu,et al.  Parallel Processing Method for Airborne Laser Scanning Data Using a PC Cluster and a Virtual Grid , 2009, Sensors.

[18]  Pankaj K. Agarwal,et al.  Natural neighbor interpolation based grid DEM construction using a GPU , 2010, GIS '10.

[19]  Paul Gray,et al.  GPU-based cloud performance for LiDAR data processing , 2011, COM.Geo.

[20]  Lei Yan,et al.  A filtering method for generating DTM based on multi-scale mathematic morphology , 2011, 2011 IEEE International Conference on Mechatronics and Automation.

[21]  D. Costantino,et al.  FEATURES AND GROUND AUTOMATIC EXTRACTION FROM AIRBORNE LIDAR DATA , 2012 .

[22]  F. Crosilla,et al.  POINTS CLASSIFICATION BY A SEQUENTIAL HIGHER – ORDER MOMENTS STATISTICAL ANALYSIS OF LIDAR DATA , 2012 .