3D-Sensing Distributed Embedded System for the Study of Human Kinetic Behavior
Andrés Burbano · 2016
The present study focuses on the development of an embedded, low-cost and scalable smart camera network dedicated to tracking and counting people in public spaces. In the network, each node is capable of sensing, tracking and counting people while communicating with the adjacent nodes. Each node typically uses a 3D-sensing camera positioned in a downward-view, capable of sensing and tracking people individually. Additionally, it is able to simultaneously communicate with the adjacent nodes and perform people transition between different cameras. The system saves and analyzes the people's global trajectories in order to determine space usage and people's behavior. On the other hand, we constructed a dynamic architecture that accelerates the integration process of new depth cameras to the system functionality. The design and processes definition allows the system to be scalable and easy to install. During the development of this thesis, several smart cameras and services have been implemented and released into the market.