Remote Sensing – A short introduction

by Hanna Schreier

“Man must rise above – to the top of the clouds and beyond – for only thus he fully understand the world in which he lives.”

Socrates, ~410 BC

More than 2000 years ago Socrates already envisioned earth observation from above. But before we could monitor earth from space via satellites, the development of photography, aeronautics and electromagnetism were important steps in the field of remote sensing.

The evolution of air- and spaceborne imagery was mainly driven through military applications. Only in the 1970s the civil usage of satellite data began with the launch of the environmental imaging sensor LandSat. Ever since, the launching of satellites increased exponentially.

Remote Sensing can be described as the practice of deriving information about Earth’s surface using images acquired from an overhead perspective from one or more regions of the electromagnetic spectrum reflected or emitted from the surface.

So, let’s explore the electromagnetic spectrum a little:

The electromagnetic spectrum can be defined as the set of all electromagnetic waves which can be distinguished by energy at different wavelengths and frequencies (Fig. 1). In earth observation visible, infrared and radar wavelengths are used.

Fig. 1: Frequency and Wavelength along the Electromagnetic Spectrum (Source: InductiveloadNASAEM Spectrum Properties edit, Edited by Hanna Schreier, CC BY-SA 3.0)

Remote sensing is based on the assumption that every land cover type or object has its own spectral signature. This means that the reflectance in the different parts of the electromagnetic spectrum lets you distinguish between different land cover types (Fig. 2). Vegetation for example has very low reflectance in the visible spectrum (380 – 740 nm), but has a characteristic sharp rise on the edge to the near-infrared (750-2500 nm), which is also called the “red edge”.

There are different types of earth observation data, which can be basically split into optical and radar imagery. Each of these data types has strengths and weaknesses. Radar data for example can also be collected when it is cloudy, on the other hand data is not as easily interpretable as optical data. Therefore, a combination of different datasets can help to use their respective strengths.

Remote Sensing brings great advantages to earth observation as it can cover vast expanses of land per image. Additionally, the same areas are acquired regularly, thus changes can be identified easily. Inhospitable areas pose no problems anymore.

The range of applications and end users in remote sensing is very diverse, reaching from academia or industry to governments. Additionally, the increasing availability of geospatial data as well as the enhancement of spatial resolutions build the perfect basis for even more advanced applications.

As an example, LiveEO uses cutting-edge machine learning algorithms in order to identify threats, such as vegetation, ground deformation or third-party interactions close to infrastructure networks on the basis of satellite data. Thereby we help to increase the network’s availability while saving operational costs by replacing manual monitoring services.