Photogrammetry and Structure from Motion Concepts

Center for Geospatial Analytics at North Carolina State University

Justyna Jeziorska, Helena Mitasova & Corey White

Objectives

Understand the role of remote sensing and photogrammetry in geospatial data aquisition
Describe different types of aerial photography and associated challenges for measurements
Understand why photogrammetry needs to be used to make measurements from aerial photographs
Understand Structure from Motion concepts in creating 3D models from 2D images

What is Remote Sensing?

Sensing without contact
Acquiring data from a distance
Passive sensors: sensor captures reflected rays
Active sensors: sensor sends out rays and captures the return signal (lidar, radar, sonar)

Remote sensing for mapping explained

Remote sensing, photogrammerty and GIS

Source of diagrams: Schenk (2005), modified

Photogrammetry

3-D coordinate measuring technique that uses PHOTOGRAPHS as the fundamental medium for measurement (the science of taking precise measurements from photographs);
Can be classified into two types: aerial and terrestrial (close range);
Aerial Photogrammerty was a crucial development in map making

Vertical and oblique aerial imagery

Vertical aerial photography

Jockey’s Ridge 1932

Oblique aerial photography

San Francisco Earthquake, 1906: Note the high distortion

Photogrammetric process

Geometry of aerial photograph

Why do we need to process the images?

From perspective to orthographic view

Why do we need to process the images?

Perspective and relief distortion

Orthorectification - Perspective and relief distortion

Process that removes:

geometric perspective
effects of relief displacement
optical distortions from the sensor

from a photograph or digital image

The resulting image - an orthophoto or orthoimage.

Orthorectification

Orthophoto

Photo that is geometrically corrected and has a uniform scale;
Can be used to measure true horizontal distances (depending on cartographic projection and spatial extent)

How do we get there?

Old way: analogue

Now: digital

What do we need?

Digital imagery with sufficient overlap;
(Digital elevation model to remove the relief distortion)
(Exterior orientation parameters from aerial triangulation or Inertial Measurement Unit (IMU) to remove tilt distortion);
(Camera calibration report, to transform perspective to ortho projection);
(Ground Control Points for georeferencing and to reduce distortions);
Photogrammetric processing software that utilizes collinearity equations.

Parameters 2,3,4 can be estimated from the images - see lecture on Imagery processing

1. Digital imagery

Location of the photo

Geotagging a photograph - associating a photo with a geographical location (latitude, longitude, and usually altitude)

In theory, every part of a picture can be tied to a geographic location, but in the most typical application, only the position of the sensor is associated with the entire digital image.
GPS in the camera or UAS measures location with very low accuracy.

Geotagging

5. Ground Control Points

GCP - target in the project area with known 3 coordinates (X, Y, Z or lat, long, alt)
Accurate, well-placed, and marked GCPs are essential elements for model accuracy and georeferencing

Photo Identifiable: - Any feature on the ground, - Specific (e.g., corners) - Unmovable - Not covered by vegetation - It can be surveyed later on.

Ground Control Points

Pre-marked (Panels): marking or painting figures or symbols on the ground before the UAS flies

Why Ground Control Points?

necessary for georeferencing if photos are not geotagged
improve precision of the model
important for quality control

Why Ground Control Points?

From 2D images to a 3D model

orthophoto is a 2D image
elevation data are derived as part of orthorectification process
exact camera parameters and manually identified GCPs on the images were needed to derive a DEM
Structure from Motion: automated point matching, camera parameter estimation and 3D model generation

Multiple-view geometry

Scene geometry (structure):
Given 2D point matches in two or more images, where are the corresponding points in 3D?
Correspondence (stereo matching): Given a point in just one image, how does it constrain the position of the corresponding point in another image?
Camera geometry (motion): Given a set of corresponding points in two or more images, what are the camera matrices for these views?