| Airborne LiDAR system provides geo-referenced high-resolution scanning point cloud data with additional attributes such as intensity, returning number, and RGB color information, which can be used to create 3D terrain models useful in augmented or virtual reality in which real-world landscapes are incorporated. In this report, we propose a framework based on georeferenced LiDAR point cloud data that detects building and vegetation features and reconstructs 3D surface models for each feature object automatically. The key contributions from this work are: (1) A complete system to detect man-made structures and vegetation areas automatically given a high resolution LiDAR data set; (2) A new method to accomplish fast footprint extraction and accurate 3D man-made structure surface reconstruction with real-world color information applied; (3) A new approach to delineate and reconstruct single tree canopy models from vegetation areas and applies the real-world color information. The test data sets are the real-world high-resolution LiDAR data sets sponsored by the LiDAR development department, John Chance Land Surveys, Inc., Lafayette, Louisiana. |
