Development of a sensing seat for human authentication

In this work the development of a sensing seat for human authentication is reported. Such a system can be used in all the critical scenarios where a seat is available to the human subject. In order to face the authentication task, the sensing seat was developed by means of a novel unobtrusive sensing technology. This is mainly due to two aspects: the unavailability of an existing sensing seat system for human authentication and the inadequacy of the existing sensor technology in order to address the human subject recognition as well as the integration of the sensors in the seat. Thanks to the development of a redundant sensor network, we adopted a hierarchical architecture. Three cooperating classifiers (a distance-based classifier, a KSOM and a MLP) shares the input data and supplying three different classification results. A final classifier (a weighted averager) performs the fusion of the results and supplies the final response.