Spatio-temporal event detection using dynamic conditional random fields

摘要: Event detection is a critical task in sensor networks for variety of real-world applications. Many events often exhibit complex spatio-temporal patterns whereby they manifest themselves via observations over time and space proximities. These cannot be handled well by many the previous approaches. In this paper, we propose new Spatio-Temporal Detection (STED) algorithm based on dynamic conditional random field (DCRF) model. Our STED method handles uncertainty data explicitly permits neighborhood interactions both event labels. Experiments real synthetic demonstrate that our can provide accurate near even large-scale networks.