5 Ways to Optimize Your LiDAR Data
0 likes936 views
The document outlines five strategies to optimize lidar data workflows, emphasizing the importance of simplifying data transformation, preparation, surface model creation, and utilizing visualization techniques in 3D city modeling. It also highlights the integration of FME tools and third-party applications to enhance lidar data management, including tips for effective use of these tools. Participants are encouraged to engage in community activities and explore additional resources for improving their lidar data capabilities.
5 Ways to Optimize Your LiDAR Data
- 1. 5 Ways to Optimize Your LiDAR Data
- 2. Meet the Presenters Kailin Opaleychuk Technical Support Specialist I, FME Desktop Richard Mosley Technical Support Specialist II, FME Server Jovita Chan Technical Support Specialist II, FME Desktop
- 3. Agenda ● LiDAR Overview ● 5 Ways to Improve Your Lidar Workflows 1. Simplify the process of transforming and integrating point clouds 2. Prepare your data 3. Create (and automate the creation of) surface models 4. Visualize solutions by using LiDAR for 3D City Modelling 5. Expand your toolset with FME Hub and Third Party Tools ● Q&A
- 5. What is LiDAR? ● LiDAR stands for Light Detection and Ranging ● LiDAR is used to collect and create point clouds ● Point clouds are sets of points that describe an surface
- 6. Supported Point Cloud Formats ● E57 ● Cesium 3D Point Cloud ● LAS/LAZ ● Mojang Minecraft ● Point Cloud data (PCD) ● Point Cloud XYZ (XYZ) ● Oracle Spatial Point Cloud .... and more!
- 7. Point clouds can help you produce amazing 3D visualizations!
- 8. The problem with point clouds... They can be huge. Theyʼre kinda complicated if you are new to working with them. There can be limitations. Itʼs hard to integrate them with other data. Working with them can be expensive.
- 9. FME® Integration Platform Connect. Transform. Automate. FME Desktop FME Server FME Cloud Build & Run Data Workflows Automate Data Workflows (on-premises) Automate Data Workflows (cloud) Get a free trial of FME Desktop and FME Server at safe.com
- 10. Pointing you in the right direction with FME ● Create a surface model ● Combine with other formats ● Split datasets ● Volume/point reduction ● Clip to a specific region or AOI ● Perform calculations (stats) ● Point Cloud splitting
- 11. 1. Simplify the process of transforming and integrating point clouds using FME’s Point Cloud Transformers.
- 12. Our transformers simplify the process of transforming and integrating point clouds. Combining transformers into workspaces creates powerful, repeatable solutions.
- 16. 2. Prepare your data
- 17. Data Preparation ● Clip a Point Cloud ● Noise removal ● Creating an extent or AOI ● Tiling
- 18. 3. Create and automate surface models
- 19. How FME Understands Point Clouds ● Point Cloud as 1 feature ● Each point has is own component values ● Each individual component value can be altered without it being an attribute ● Attributes from other sources can be added as components.
- 20. Understanding Components Vs. Attributes Attributes Components Each point has its own component values
- 21. Clip, Tile, and Thin Especially when designing workflows use smaller sections to speed up development
- 22. Demo
- 23. DEM Parameters Set appearance on: Front Side Texture Mapping Type: Top Down Georeferenced
- 24. Digital Surface Model Or any other classification model….
- 26. Demo
- 27. Automate and Self Serve Self-Service ● Geometry Picker ○ https://community.safe.com/s/article/using-the-geometry-picker-in-fme-server Speed up ● Dynamic Engines ○ 1 engine can process 1 workspace at a time. So if you need some performance gains split your workflow and process smaller amounts of data on more engines. ○ https://community.safe.com/s/article/getting-started-with-dynamic-engines
- 28. 4. Visualize solutions by using LiDAR for 3D City Modelling
- 29. Demo: Creating 3D City Models ● Goal: Visualize the city as an accurate 3D model to ensure developments are integrated well in their neighborhoods and to inform future decisions ● Obstacles: Large datasets, different formats, which can be difficult to work with. ● Solutions: Use FME to extract needed information from LiDAR, integrate with other formats like PNG textures and SketchUp models, and apply that to a 3D City Model
- 30. Source Data ● AutoCAD (DWG) - building outlines ● Point Cloud (LAS) ● MapInfo (MiTAB) - Park polygons ● GeoTIFF (TIFF) - 4 orthoimages
- 31. Source Data Continued ● 3D SketchUp (SKP) models ● PNG/JPEG images - for wall textures Wall texture from: https://pxhere.com/en/photo/1421981 SketchUp Tree from: https://3dwarehouse.sketchup.com/ model/e18aebbf-2859-47f1-805c-8a4f 946f5893/Skatter-Tree-02
- 32. Goals for this workspace 1. Create 3D buildings using LAS and DWG building outlines 2. Add PNG and orthophotos (GeoTIFFs) as textures on 3D models 3. Filter data to get point features of trees and replace them with instances of a 3D tree model
- 33. Demo
- 35. Workspace Highlights 1. Create 3D buildings using point cloud and vector building outlines https://community.safe.com/s/article/point-cloud-to-3d-terrain-model-with-build ings-dwg 2. Add image and orthophotos as textures on 3D models 3. Filter data to get point features of trees and replace them with instances of a 3D tree model https://community.safe.com/s/article/derive-a-boundary-from-a-lidar-point-cloud https://community.safe.com/s/article/creating-and-using-geometry-instances
- 36. Tips and Tricks: Methods ● Similarities to raster processing ○ Use feature caching strategically ○ Tiling large datasets ● Knowing how FME works with point clouds ○ FME treats a point cloud as a single entity ■ For many operations, use Point cloud specific transformers (attributes vs. components) ○ Writing it out to a format that doesnʼt support point clouds means youʼre actually writing a bounding box ■ Coerce to multipoints, surface, or raster
- 37. Tips and Tricks: Transformers 3D Geometry ● TINGenerator - creates surface models ● Extruder - extrudes outline features (ex. buildings) Applying textures ● AppearanceSetter - to add image textures, pay attention to the Texture Mapping Type parameter
- 38. 5. Expand your toolset with FME Hub and Third Party Tools
- 39. PointCloudSpatialThinner This transformer spatially thins point clouds by the user-defined resolution. IFMEPointCloud: 4377463 Points IFMEPointCloud: 4025219 Points
- 40. Demo
- 41. PointCloudStatsRasterizer Replaces a point cloud with a raster which represents statistical values of a component of the source point cloud.
- 42. Demo
- 43. LAStools LAStools.lasground: for bare-earth identification: it classifies the LiDAR points into ground points (class = 2) LAStools.lasheight: computes the height of each LAS point above the ground. LAStools.classify: classifies buildings and high vegetation (i.e. trees). LAStools is produced by Rapid Lasso GmbH
- 44. 1. Read in LAS file using the ASPRS LiDAR Data Exchange Format (LAS) Reader
- 45. 2. Remove existing classification
- 46. 2. Use QuickAdd to add LAStools.ground
- 47. 2. Now that ground is classified, add LAStools.lasheight
- 48. Tips and Tricks: LAStools ● Disable feature caching ● If tools are not running properly and or erroring without much indication, try purging your temp file (FME Workbench > Tools > Purge Temporary Folder) ○ LAStools.lasheight_Terminator (TestFactory): LAStools.lasheight_Terminator: Termination Message: 'Translation Terminated. lasheight.exe wasn't executed successfully' ○ A fatal error has occurred. Check the logfile above for details ● If possible, shorten the file path as much as possible in the transformers ○ C:devLAStoolsLAStools-binbinexecutables.exe
- 49. Summary
- 50. Working with LiDAR data shouldn’t crash your computer. With the right tools and knowledge, you can feel confident integrating LiDAR with the rest of your data and get the most out of LiDAR data.
- 51. Improve your LiDAR workflows by: 1. Simplifying the process of transforming and integrating point clouds using transformers. 2. Preparing your data. 3. Automate and Scale with FME Server 4. Visualizing solutions by using LiDAR for Surface Models and 3D City Modelling. 5. Expanding your toolset with FME Hub and Third Party Tools.
- 52. Claim Your Community Badge Get community badges for watching webinars! fme.ly/WebinarBadge Todayʼs Code: SCGAM
- 53. Q&A
- 54. The Peak of Data Integration 2022 UC August 24-26, 2022 Vancouver, Canada Register now
- 55. Check out our upcoming & on-demand webinars: safe.com/webinars
- 56. Thank you! Download FME 2022.0 Beta Free Trial | Upgrade Contact us info@safe.com Connect with us in the Community Connect with us for more FME Please share your feedback with us through the webinar survey!