Copernicus data are becoming tremendous in volume & quality. Copernicus services are focusing on professional services, like the Emergency Management Service whose aim is to provide information on major events to the rescue and security forces. In order to raise awareness of the Copernicus services to the EU citizens, there is a need to provide added value data to the Media Industry, rendering them to major demonstrators of the success of Copernicus to the citizens.
In this direction, the impact of the SnapEarth project will be enhanced through services for combinatorial EO data retrieval, and the implementation of data correlations between EO and news information.
In particular, the EarthPress (https://snapearth.eu/services/earth-press) pilot aims to provide services to editors and journalists, allowing them to enrich the content of their publications and articles with EO data, by providing an almost ready to be published article.
The focus of EarthPress is disasters’ reporting (e.g. floods, wildfires, earthquakes, explosions, tornados, hurricanes, volcano eruptions, general disaster). A report with the before/after depiction of a disaster is generated, accompanied by rough statistics on the impacted area (e.g. surface, type of land cover/ land use affected) and EO images depicting the affected areas. This report is enriched with data collected from citizens journalism posted on social networks, including textual information, multimedia, processed EO images and numerical data concerning the damage that has occurred due to a disaster, e.g., the damage in percentage occurred in a building after an earthquake (general disaster).
EarthPress is a web-based platform that that supports the Media & Press Industry with a set of valuable tools and services for Earth-related event reporting. It is a service that:
- provides access to multimedia data from multiples sources;
- detects automatically breaking news related to disasters;
- distinguishes real from fake news;
- extracts and present useful information and statistics from geospatial data;
- provides changes in land cover in the form of information layers;
- synthesizes ready to print AI generated article tailored to the user’s profile;
- provides all the above in a single platform;
- reduces time needed for publishing news articles.
Figure 1: User Interface of the EarthPress platform – Query on disasters by filtering (type of disaster, date, location)
Figure 2: EarthPress provides a list with articles and news from Twitter that are related to the specific event
Figure 3: EarthPress also generates a list with images related to the event that are available online.
Figure 4: Through the Image Processing module the user is able to retrieve EO images from the area of interest that depict the area before the disaster occurred, after the passing of the disaster and an EO image that is overlayed with the delineation of the affected area.
Special focus has been made within EarthPress on the analysis of EO images through the use of the corresponding Image Processing AI models. Through the image processing processes performed, satellite images referring to the area of a disaster are collected and processed aiming to identify the affected area. For each affected area identified, the corresponding processed image, including an overlay depicting the affected area and its density, is provided as output along with statistical data about the disaster.
Figure 5: Selection of a specific event, i.e. fire in Lesvos island, Greece with results that include EO analysis. Overlay with the affected area is presented in the map.
Figure 6: Additional information such as the total size of the affected area, the Percentage of the land cover affected (based on the CLC nomenclature) and the uncertainty of the calculations are available for each processed EO image through the EarthPress platform.
Concerning the almost ready to be published article provided by the platform, EarthPress allows editors and journalists to spend their time on trimming and finalizing the generated article, therefore reducing the time spent on writing an article.
Figure 7: The users will have at their disposal the available options for saving and downloading the article as a pdf file.