This project is created and developed for the I3 Project at the University of Salzburg.
This project aims at developing a Google Earth Engine Application that allows to visualize and analyze Sentinel-1 data for humanitarian applications. A first focus is set on implementing functionalities for dwelling detection.
The last 15 years have seen an increase of applications where geospatial technologies and Earth Observation data are used to support operations of humanitarian NGOs. The main benefit thereby is the possibility to obtain reliable information of crisis areas that are usually non-/difficult to access. Currently, the majority of information is extracted from optical satellite imagery although radar data can complement the information retrieval due to its inherent advantages concerning data acquisition in cloudy areas and during nighttime. Although more difficult to interpret, is has shown its benefits in dwelling detection and flood detection. Therefore, this project aims at creating an interactive Google Earth Engine App for dwelling detection that allows the user select Sentinel-1 data from different time steps/periods, possibly calculating mean images, and visualizing the results in split screen images or false color composites. The user should also be able to select thresholds to identify candidate areas where possible dwellings have appeared and digitize them into a vector dataset which in the end can be exported for further analysis. Possible additional features are built-in classification routes based on in-app sample selection or result optimization with additional data from the GEE (Google Earth Engine) data catalog. In the end, an easy to share, intuitive application tool should be created that allows users from humanitarian organizations to quickly get radar-based information on crisis areas of their choice.
Keywords: Sentinel-1, Earth Engine, App, Earth Observation, Humanitarian application
- Working Google Earth Engine App
- User-friendly and intuitive UI
- lowering entry barrier for radar-based information retrieval
The current project status is shown here
By 30.06.2022 the project in context of the I3 project is finished and closed. All relevant documents are handed in and/or available in this repo.
The App can be accessed here Have fun exploring radar!
For feedback, improvement ideas, comment, critique and everything else please email me (s1079105@stud.sbg.ac.at).
@Niklas Jaggy