Extracting stay regions with uncertain boundaries from GPS trajectories: a case study in animal ecology

摘要: In this paper we present a time-aware, density-based clustering technique for the identification of stay regions in trajectories low-sampling-rate GPS points, and its application to study animal migrations. A region is defined as portion space which generally does not designate precise geographical entity where an object significantly period time, spite relatively short periods absence. Stay can delimit example residence animals, i.e. home-range. The proposed enables extraction represented by dense temporally disjoint sub-trajectories, through specification small set parameters related density presence. While work takes inspiration from field ecology, argue that approach be more general concern used perspective different domains, e.g. human mobility over large temporal scales. We experiment with on case study, regarding seasonal migration group roe deer.