Interactive Reception Desk with Face Recognition-Based Access Control

In this paper, we present an automatic reception desk demonstration system which exploits facial recognition for securing access to facilities and providing innovative services. Facial characteristics are attractive biometric traits in the addressed application for the non-intrusiveness and the easiness of their acquisition through digital cameras. The system is equipped with a camera to capture images of visitors, which are then processed in order to allow or deny the access to a certain area. Face and eye detection techniques are exploited to automatically extract the facial region where to perform face recognition by means of a regularized linear discriminant analysis (R-LDA). The benefits of exploiting color information for recognition have been taken into account by applying the R-LDA approach to image color components. The system has been tested on standard datasets and on an internal database collected over a period of six weeks.