Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Land Use Detection

218 views

Published on

Published in: Technology, Business
  • Be the first to comment

  • Be the first to like this

Land Use Detection

  1. 1. Zhaoying Wei GEOG 8350 Comparison of Land use detection in Niagara Falls using LIDAR and orthoimagery
  2. 2. Outlines  Goal  Area of Interest  Data  Method and Result  Conclusion and Future Work
  3. 3. Land use detection orthophoto LIDAR Goal
  4. 4. AOI
  5. 5. Data Data • point cloud Lidar data • orthoimagery Tools •LAStools •Lidar Analyst •ArcGIS 10.0 • ArcGIS 10.0 • Erdas • ENVI
  6. 6. Method (LIDAR)  Extract Bare earth • strip the existing classes from the LAS file and set the coordinate system and projection (noclass.las) • classify the ground points (ground.las) • create the only ground points (groundonly.las) • generate DEM of groundonly .las • extract bare earth DEM from groundonly DEM
  7. 7. Method (LIDAR) From From raw LAS  Extract Bare earth
  8. 8. Method (LIDAR) Extract buildings footprints • compute the height above the ground • Classify trees and Buidlings (treeNbuilding.las) • generate new bare earth for building and trees extraction
  9. 9. Extract building footprints • point cloud building extraction Method (LIDAR)
  10. 10. Extract building footprints • edit building Method (LIDAR) Before After Remove building Add to building Merge building
  11. 11. Before After Extract building footprints Method (LIDAR) Create new building and close holes Final footprints
  12. 12. Extract tree • point cloud tree extraction • visually edit forests and trees Method (LIDAR)
  13. 13. Method (Orthophoto) Supervised classification • Create the signature file and AOI for each class • Combine the signature file of all classes
  14. 14. Method (Orthophoto) Reclassify • manually digitize AOI • assign new class value to water
  15. 15. Conclusion  LIDAR: • vector results consistent with real terrains • limited classes Orthoimage • numerous classes • low quality and accuracy, water
  16. 16. Future work Measure accuracy of the classification result  improve the quality of signature  better water detection  object-based classification Fusion of LIDAR and aerial image

×