Real time tracking and monitoring of human behavior in an indoor environment

Abstract

This chapter reports the development of a real time 3D sensor system and a new concept based on space decomposition by encoding its operational space using limited number of laser spots. The sensor system uses the richness and the strength of the vision while reducing the data-load and computational cost. The chapter presents the development and implementation of an intelligent 3D Fiber Grating (FG) based vision-system that can monitor and track human being status in real time for monitoring purposes to support wide range of applications. The 3D visual sensor is able to measure three-dimensional information with respect to human, objects and surrounding environment. The sensor system consists of a CCD camera, a laser spot array generator (constitutes: laser diode and driver, lens, fiber gratings and holder), and a processing unit with alarm facilities and interfacing capabilities to a higher-level controller and decision-making along with a user-friendly interface. The system works by projecting a two-dimensional matrix of laser spots generated through two perpendicularly overlaid layers of FGs. Then, the spot array generator projects the laser spot array on the front scene within the active view of the CCD camera and the reflected laser spots from the scene play an essential role in detecting and tracking targets. It is possible to adjust or translate the position of the laser spot generator with respect to the CCD camera to have better and balanced resolution. In addition, it is possible to rotate the FG in order to obtain a better laser spot pattern that can facilitate processing and enhance accuracy. Furthermore, multi-laser spot generators can be configured with one CCD camera to widen the operational coverage of the developed sensor system. This chapter introduces and illustrates the structure of the developed sensor system, its operational principles, performance analysis and experimental results. © 2008 Springer-Verlag Berlin Heidelberg.