escience2019 has ended
Friday, September 27 • 11:30am - 12:00pm
AdaptLidarTools: A Full-Waveform Lidar Processing Suite

Sign up or log in to save this to your schedule and see who's attending!

Ravi Shankar (Boise State University), Nayani Ilangakoon (Boise State University), Aaron Orenstein (Treasure Valley Math and Science Center), Floriana Ciaglia (Boise State University), Nancy Glenn (Boise State University), and Catherine Olschanowsky (Boise State University)

AdaptLidarTools is a software package that processes full-waveform lidar data. Full-waveform lidar is an active remote sensing technique in which a laser beam is emitted towards a target and the backscattered energy is recorded as a near continuous waveform. A collection of waveforms from airborne lidar can capture landscape characteristics in three dimensions. Specific to vegetation, the extracted echoes and echo properties from the waveforms can provide scientists its structural (height, volume, layers of canopy, among others) and functional (leaf area index, diversity) characteristics. The discrete waveforms are transformed into georeferenced 2D rasters (images). The georeferencing orients the raster on a map and allows scientists to correlate field-based observations for validation of the waveform observations. AdaptLidarTools provides an extensible, open source framework that processes the waveforms and allows multiple processing methods.

AdaptLidarTools is designed to explore new methods to fit full-waveform lidar signals and to maximize the information in the waveforms for vegetation applications. The toolkit explores First Differencing, complementary to Gaussian fitting, for faster processing of full-waveform lidar signals and for handling increasingly large volumes of full-waveform lidar datasets. AdaptLidarTools takes approximately 30 min to derive a raster of a given echo property from a raw waveform file of 1 GB size. The toolkit is designed to generate the first order echo properties such as position, amplitude, pulse width, and other properties such as rise time, fall time, backscattered cross section and others that current proprietary and open source tools do not generate. The derived echo properties are delivered as georeferenced raster files of a given spatial resolution that can be viewed and processed by most remote sensing data processing software.


Ravi Shankar

Boise State University

Friday September 27, 2019 11:30am - 12:00pm
Macaw Room