AdaptLidarTools: Full Waveform LiDAR Data Analysis
College of Engineering
Department of Computer Science
Dr. Catherine Olschanowsky
AdaptLidarTools is an open source package that processes full waveform LiDAR data. LiDAR is a remote sensing method that captures landscape characteristics from LiDAR units mounted on aircraft. The data is delivered as sets of waveforms. Interpreting the data requires transforming discrete waveform data into geo-referenced 2D rasters. A variety of data characteristics can be captured in the raster files; the current tool captures the maximum elevation observed at each meter in the scanned domain. Furthermore, processing a 1GB waves file (*.wvs) which previously took over a day is being processed by the toolkit, using Gaussian fitting, in approximately 33 minutes. The toolkit is designed to explore alternative methods to fit the waves. For example, first differencing in lieu of Gaussian fitting for faster processing of data. This poster presents initial results that indicate the first differencing results are comparable for specific landscapes. This approach takes approximately 5 minutes to process the same input that took Gaussian fitting 33 minutes. Further work is required to determine the statistical significance of the differences between these two methods, whereby future iterations of this toolkit can effectively balance precision and speed.
Shankar, Ravi; Orenstein, Aaron; and Ciaglia, Floriana, "AdaptLidarTools: Full Waveform LiDAR Data Analysis" (2019). 2019 Undergraduate Research and Scholarship Conference. 197.