Overview of the ImageCLEF 2017 Population Estimation (Remote) Task

Helbert Arenas (1) , Md Bayzidul Islam (2) , Josiane Mothe (3, 4)

1 IRIT-MELODI - MEthodes et ingénierie des Langues, des Ontologies et du DIscours
2 TU Darmstadt - Technische Universität Darmstadt - Technical University of Darmstadt
3 IRIT-SIG - Systèmes d’Informations Généralisées
4 UT2J - Université Toulouse - Jean Jaurès

Estimating population has many applications as planning disaster responses, communication infrastructure, or development activities. In 2017, ImageCLEF Lab introduced a new pilot task: the Population Estimation (or Remote) task which aims at estimating the population of an area of interest by exploring Copernicus earth observation data (i.e. free Sentinel-2 satellite images). In line with the goal of FabSpace 2.0 project, a European Unions initiative to bring Geo-Enthusiast together around the 6 European universities in collaboration with Business Incubation Centers, participated to this challenge. This paper presents the results that were obtained by the participants as well as a brief summary of some of the approaches that were used.

Mots clés: Information systems; Information retrieval; Image retrieval; Copernicus data; Automatic population estimation; FabSpace 2.0; CLEF

Domaines: Théorie de l'information [cs.IT]; Recherche d'information [cs.IR]

Helbert Arenas, Md Bayzidul Islam, Josiane Mothe. Overview of the ImageCLEF 2017 Population Estimation (Remote) Task. International Conference of the CLEF Association, CLEF 2017 Labs Working Notes (CLEF 2017)), Sep 2017, Dublin, Ireland. pp. 1-13. ⟨hal-01912784⟩