Field walkthrough — Surveying 120 acres with a drone
The hook
Photogrammetry derives 3D measurements from 2D photos — by capturing the same feature from multiple angles and reconstructing depth from parallax. Drones turned a once-rare specialty into something every surveyor has access to. Remote sensing is the broader field: satellites, lidar, multispectral imagery, and the analytics built on top of them.
Memorize these
Concepts that show up on the exam
Stereo / multi-view photogrammetry
Reconstructing 3D coordinates from images taken at multiple positions. Parallax + known camera positions + matched features → 3D point cloud.
Structure from Motion (SfM)
Modern photogrammetry that simultaneously solves camera positions AND scene geometry from a set of overlapping images. The technique behind Pix4D, Metashape, OpenDroneMap, etc.
Ground Sample Distance (GSD)
The real-world size of one pixel in the photo. GSD = (sensor pixel size × flight altitude) / focal length. Smaller GSD = finer detail. Drives flight planning.
Forward / side overlap
Drone mission parameters. 75-80% forward (between consecutive shots) and 60-70% side (between adjacent flight lines) is typical. Less overlap = gaps; more = wasted flight time.
GCP (Ground Control Point)
A surveyed point with known XYZ that's visible in the photos. Used to constrain the photogrammetric solution to real-world coordinates. 4-8 GCPs is typical for a small site.
Orthomosaic
A rectified, seamless aerial image where distances on the image equal distances on the ground. Output of every modern photogrammetric workflow.
LiDAR
Light Detection And Ranging — pulses of laser light measure round-trip distance. Active sensor (works at night). Penetrates vegetation; produces a much "cleaner" point cloud than photogrammetry.
Multispectral / hyperspectral
Imagery that captures multiple bands beyond visible light (NIR, red-edge, etc.). Used for vegetation health (NDVI), classification, and change detection.
| Sensor / method | Cost | Data type | Best for |
|---|---|---|---|
| Drone photogrammetry | $ | Point cloud + orthomosaic | Site topo, volume, visual record |
| Drone LiDAR | $$$ | Dense point cloud (cm) | Vegetated sites, power lines, complex 3D |
| Manned aerial mapping | $$$ | Wide-area orthos + DEM | County-wide topos, infrastructure corridors |
| Satellite imagery (free: Landsat / Sentinel) | $ (often free) | Multispectral raster (10-30 m) | Regional planning, change detection |
| Terrestrial laser scanning | $$ | Ultra-dense indoor/façade scans | As-built BIM, historic structures, complex geometry |
Don't fall for these
What trips people up
Skimping on GCPs
Without sufficient GCPs, the photogrammetric block can drift in absolute position by meters. Internal accuracy stays good, but the whole model is shifted. Plan for 4-8 GCPs distributed around the site.
Confusing accuracy with resolution
1 cm GSD doesn't mean 1 cm accuracy. Accuracy is GCP-driven (typically 2-5x the GSD); resolution is camera/altitude-driven. Don't over-promise based on pretty-looking pixels.
Vegetation confusion in photogrammetry vs. LiDAR
Photogrammetry sees what the camera sees — the top of the tree canopy. To get bare earth under trees, use LiDAR (which penetrates) or skip those areas. Don't deliver a "DTM" that's actually a DSM.
Test yourself
How well did it stick?
A quick 5-question check on Photogrammetry and Remote Sensing. See where you stand and what to review.