After data is collected from the field, it gets “processed.”  This term gets thrown around a lot as it broadly describes a wide number of steps to produce the requested deliverable.  Processing often takes 1 to 20+ times the effort than the  effort it took to collect the raw data in the field.  The image above shows the difference between several common output steps.  DSM – Digital Surface Model, includes vegetation while a DTM – Digital Terrain Model removes all items not considered ground.

Basic field collection and processing typically generates two products:  an orthomosaic image and point cloud.  After this step, further processing might include:

Point Cloud:  Whether using SfM or LiDAR, processing point clouds can include:

• Project geo-referencing, Quality Assurance and adjustment
• Orthomosaic tiling and compression
• Point cloud classification
  • Algorithmic
  • Manual
• Point cloud density reduction
• Surface model generation
  • LandXML
  • GeoTIFF
• Extraction of features, such as power lines, tree sizes, etc
• Extraction of contours
  • Cartography or engineering grade
• Planimetric extraction of standard 2-D linework
• Export into a wide number of formats