Cloud-Based LiDAR Visualization and ExploitationASPRS - 2015Patrick Collins

LiDAR is a Growing IndustryExpected to grow to over 550 million dollars by 2018* Reduced cost for acquisition and analysis More businesses getting into LiDAR development and servicesLiDAR is uniquely solving complex problems across industries Advanced Driver Assistance Systems (ADAS) Offshore Wind Measurement Forest and Crop Assessment / Management Urban Planning and Development* LiDAR Market by Components (INS, Laser, GPS/GNSS, Camera, MEMS), Product Types (Airborne, Mobile, Terrestrial),Applications (Corridor Mapping, Forestry, Mining, Topographic Surveying, Volumetric Mapping) - Global Forecastsand Analysis 2013 – 2018

Web GIS is a Growing IndustryCompanies are building robust online visualization and analysis communities Google Earth Visualization of imagery, building models Esri Sharing of data layers and maps Exelis Advanced web-based analytics

The Art of the PossibleConsumers of web GIS want easy and intuitive solutions Many non-traditional users Apps and interfaces should be easy to use Making something possible does not necessarily make it easy

What we’ve doneCreated a WebGL viewer that consumes LiDAR point clouds through a browser Visualization of streaming LiDAR from ENVI Services EngineEnabled automated building feature extraction via http://REST protocols FX routines pulled from ENVI LiDAR and enabled as Services Engine tasks

Basic ArchitectureENVI Services Engine provides data streaming and analysis capabilities Ingest and display of multiple data modalities LiDAR-specific data ingest and preparation Hosting of analysis capabilities ENVI IDL C JavaWebGL Viewer consumes streaming point clouds andenables analysis calls Provides the user experience Leverages http:// Rest calls to call LiDAR analysis

LiDAR Provides Unique ChallengesCloud-based visualization and exploitation of LiDAR is different than most traditional GISmodalities Size of LiDAR datasets LiDAR data needs to be massaged prior to dissemination User wants desired information without having to do the ‘heavy lifting’ User wants fast resultsWe’ll look at two aspects of cloud-based LiDAR from two angles Visualization Make it Possible Make it Easy Exploitation Make it Possible Make it EasyData courtesy of Merrick

Web-based LiDAR visualization – make it possibleWhat are some technical considerations when visualizing LiDAR on the web? Size of LiDAR datasets Data preparation Pre-processing IDL task bins data into a quadtree structure Choosing the right viewer technology JavaScript / WebGL

Web-based LiDAR visualization – make it easyWhat are some UI/UX considerations when visualizing LiDAR on the web? Don’t stream the entire dataset! Load resolution levels based on user perspective for better performance Make the interface intuitive Zooming, panning, selecting data, and running analysis Allow users to upload their own data for visualization / analysis Automatic data prep

Web-based LiDAR exploitation – make it possibleWhat are some technical considerations when analyzing LiDAR on the web? Requires data subset, data URI, and any required parameters Coordinates sent to the LiDAR task on the server Building extraction task written in IDL and leverages the ENVI LiDAR API Extracted features saved as shapefile and streamed via the server

Web-based LiDAR exploitation – make it easyWhat are some UI/UX considerations when analyzing LiDAR on the web? Enable the user to select a subset of the data Simple buttons for clearing selection and extracting buildings Future improvements - extraction status, ability to download shapefile, more tasks

What does this all mean?Web-based LiDAR visualization and exploitation will help drive the growth of theindustry The ability to stream and analyze LiDAR point clouds via the web is a reality ENVI Services Engine combined with the 3D Web Viewer Organizations can create simple applications that leverage server technology to display and exploitLiDAR point clouds The ENVI LiDAR API contains automated extraction tasks for buildings, elevation data, power lines,trees, and more Custom routines can be designed to extract almost anything from a LiDAR point cloud The key becomes designing user interfaces that are simple and that solve specific problems withina specific industry