Navigation in Dense Human Crowds Using Smartphone Trajectories and Optical Aerial Imagery

In this paper, we propose a navigation system for smartphones which enables visitors of very large events to avoid crowded areas or narrow streets and to navigate out of a dense crowd quickly. Therefore, two types of sensor data are integrated. First, optical images acquired and transmitted by an airborne camera system are used to compute an estimation of a crowd density map. For this purpose, a patch-based approach with a Bag-of-Visual-Words Framework for texture classification in combination with an interest point detector and a smoothing function is used. Second, the GPS location information and the current movement speeds of the visitors are gathered via a smartphone app and are afterwards analyzed to enhance the final people density estimation. The combined density information is afterwards used for a least-cost navigation. Two possible use cases are presented, namely i) an emergency application and ii) a basic navigation application. A prototypic implementation of the complete system is conducted as proof of concept