Lesson 1, Topic 1
In Progress

Cloud Processing

Learning objectives of this topic

  • About need of cloud processing tools for remote sensing applications
  • Platforms for cloud processing
  • First hands-on experiences for different platforms

When we talk about the processing of remote sensing data, hardware is usually used at the user’s location. However, with increasing amounts of data being used (e.g., for extensive time series analysis), cloud processing technologies become more and more important.

Why use cloud processing?

Local desktop environments with adequate hardware are a good solution to process remote sensing imagery. Nevertheless, it has its limitations, and they have become more and more obvious in recent years.

What if you want to get your hands on the entirely available Landsat archive?
Where would you store all the data you need for this analysis?
How can you quickly overlay your time series with another extensive time series?

More data and more complex processing algorithms pose heavy demands to hardware requirements of personal computers. Memory, CPU as well as GPU power (in theory) need to be upgraded frequently to keep with the technical developments in remote sensing science. In many parts across the globe, this is a limiting factor. The solution for this problem can be the use of cloud resources where you can apply your algorithms to any data source that is available on the respective server. Let’s have a look at platforms that can be used for free to process and/visualize remote sensing data.

Google Earth Engine

Google Earth Engine (GEE) is a cloud-based platform that is probably the processing service mostly used in remote sensing, when it comes to decentralized image analysis. GEE allows you to visualize data, perform any geospatial algorithm on it on-the-fly and download products from their website. It is free of charge and is able to conquer any tasks related to remotely sensed imagery computation. You can also upload your own data sets and share them with others. Browse through extensive data catalogues and find the right data set for you!

GEE provides a JavaScript and a Python API for users to access and work with the data. Due to the large user community you can find plenty of code snippets and tutorials, such as the one below, that will make your start in GEE much easier.


By loading the video, you agree to YouTube’s privacy policy.
Learn more

Load video

Try Google Earth Engine and get your hands on RS data

Another nice feature, you can easily code in GEE are GEE Apps. This allows you to put your research into an interactive website, that others (e.g. non-scientists) can then more easily use and interpret. This way, you can lower the hurdles, people outside of the remote sensing context need to take to understand the results of your work. If you need inspirations on what your personal GEE App could look like, take a look at the website below, created by Philipp Gärtner.

Try Google Earth Engine here: GEE.


SEPAL is an initiative developed by the Forestry Department of the United Nations Food and Agriculture Organization (FAO). The entire source code of this platform is open under MIT license. Contrary to GEE SEPAL is an interface where different strings come together. SEPAL utilized GEE and Amazon Web Service (AWS) for data processing and retrieval of products. This makes the platform flexible. It is accessible through a web portal, more user-friendly than GEE but less functional (users themselves cannot create the diversity of functions). Unfortunately, SEPAL does not allow users to share their workflows, which would make it even more useful. Nevertheless, it serves as a great tool for visualization and interpretation of time series information across the globe.

Try the SEPAL platform here: SEPAL.


By loading the video, you agree to YouTube’s privacy policy.
Learn more

Load video

SEPAL presentation and introduction

Sentinel Hub EO Browser by Sinergise

The EO Browser makes it possible to browse and compare full resolution images from all the data collections that are processed by Sinergise (Sentinel-1,-2,-3,-5P; Landsat 1-5,7,8; Envisat, MODIS, DEM, Copernicus layers, Proba-V & GIBS). You simply go to your area of interest (AOI), select your desired time range and additional satellite specifications (e.g., cloud coverage for optical data or orbit direction for SAR data), and inspect the resulting data in the browser. Try out different visualizations or make your own, download high-resolution images and create time lapses. Note that the EO-Browser is a tool mainly dedicated to visualization of data and statistics, rather than creating your own extensive scripts.

Sentinel Hub services are providing long-term analysis in an efficient way. The EO Browser serves as a showcase of Sentinel Hub functionality and is free of charge. It comes with a graphical user interface (GUI), to make these features available to just about anyone. In order to keep up with the latest developments, that are taking place at Sinergise’s labs can be found in the Medium blog posts.


By loading the video, you agree to YouTube’s privacy policy.
Learn more

Load video

Daniel Thiex on the Sentinel Hub functionalities