3.4 Photogrammetry

3.4.1 Introduction to Photogrammetry
According to the American Society of Photogrammetry and Remote Sensing, Photogrammetry is the “art, science and technology of obtaining reliable information about physical objects and the environment through the process of recording, measuring and interpreting photographic images”. The word photogrammetry is derived from two Greek words: photography (phos/φῶς meaning light and gramma/γράφω meaning draw/write) and metrein, meaning measurement. In other words, photogrammetry is about deriving measurements from photographs. More specifically, we can measure:
  • The orientation of the camera, i.e. to know where exactly was the spot where an image was taken from;
  • The location of an object/scene or the 3D reconstruction of an object/scene by triangulating the information on multiple images (we will learn more on this in the following lesson);
  • Recognise the object/scene;
  • (Semantically) Interpret the content of an image and automatically generate a description of the scene.


Why Photogrammetry?

An obvious question to ask is what is the value of photogrammetry in comparison to an aerial vertical photo. Why do we need to apply a seemingly complicated process to a photograph in order to e.g. create a map based on this, and why can't we generate a map by just tracing over an image? 

Aerial photography, either using a satellite or an aircraft, presents large geometric distortions due to various factors, such as the camera and lens used, the tilting of the aircraft, the angle of view etc. This means that any measurements derived from these photographs will not correspond to the real-world measurements. Therefore, applying photogrammetry to these images is a necessary step is order to rectify such distortions and produce reliable orthoimages from which we will be able to derive accurate measurements. 

Photogrammetry is not a field; however, it draws from the theory and practice of many fields, including, cartography, mathematics, computer graphics, image processing, computer vision, robotics, and remote sensing. It was originally conceived as a method suited for aerial surveying with the first aerial photography captured from a plane in 1908, a few years after the invention of the airplane by the Wright brothers. Nowadays, with the rapid advancement of software and hardware, photogrammetry can be applied to any source of imagery (aerial or terrestrial) and at a wide range of scales (from object to landscape) to measure coordinates, distances, heights, areas, and volumes that will then be used to generate topographic maps, digital elevation models (a 3D representation of a terrain’s surface), orthophotos (photograph that has been corrected to eliminate distortions and differences in scale), and 3D models. Below you can see some of the key figures and events in the history of photogrammetry from its invention until the present day. 





References

Videos