Este site utiliza cookies para lhe proporcionar uma melhor experiência de utilização. Ao navegar aceita a política de cookies.

ExtremeCGI - Monitoring Extreme Events integrating Crowdsourced Geographic Information and Real-time Sensor Data

Title: ExtremeCGI - Monitoring Extreme Events integrating Crowdsourced Geographic Information and Real-time Sensor Data

Principal investigator: Cidália Fonte

Research team: José Paulo Elvas Duarte de Almeida, Alberto Cardoso, João Porto Albuquerque, Peter Mooney, Vyron Antoniou, Jacinto Estima

Dates start/end: 10-2016 / 03-2018


Information is the base for decision-making and the quality of information available is crucial towards good decisions. In particular, the amount and diversity of geospatial data currently available is huge and constantly growing. Although this gives a great opportunity for better decisions, there are also major challenges in integrating diverse sources of data with different characteristics. The main challenge raised is to effectively mine and integrate data so to extract relevant information, especially within a very short time window. In this context, the main aim of the project is to develop an information system capable of integrating diverse sources of geospatial data in such a way that makes it easily available to authorities in order to assist them in risk assessment and mitigation in case of extreme events. 

Types of data to be integrated include: 1) Crowdsourced data gathered by citizens for various purposes (e.g. geo-tagged photos from social networks); 2) Environmental data from in-situ physical sensors (e.g. water gauges); 3) Event specific data collected by citizens using smartphone apps developed for the aim; 4) Other sources of geospatial data, including remotely sensed imagery and derived products. 

The integration of diverse sources of data enables to: 1) Perform automatic validation procedures so to assist relevant data mining, using, for instance, the geographical location; 2) Increase the exploitation of crowdsourced data capabilities to provide useful information in real-time, or near real-time – e.g. quick identification of important information to assist mitigation operations or collect data about an occurring event; 3) Combine crowdsourced data with physical measurements, so to enable data validation and their quality assessment.  The development of such system will provide authorities with valuable information, both in quantity and quality, especially in real-time situations where decisions that directly affect people’s life need to be taken quickly.