A real-time system for recognizing faces in a video stream provided by a surveillance camera was implemented, having real-time face detection. Thus, both face detection and face recognition techniques are summary presented, without skipping the important technical aspects. The proposed approach essentially was to implement and verify the algorithm Eigenfaces for Recognition, which solves the recognition problem for two dimensional representations of faces, using the principal component analysis. The snapshots, representing input images for the proposed system, are projected in to a face space (feature space) which best defines the variation for the face images training set. The face space is defined by the ‘eigenfaces’ which are the eigenvectors of the set of faces. These eigenfaces contribute in face reconstruction of a new face image projected onto face space with a meaningful (named weight).The projection of the new image in this feature space is then compared to the available projections of training set to identify the person using the Euclidian distance. The implemented system is able to perform real-time face detection, face recognition and can give feedback giving a window with the subject's info from database and sending an e-mail notification to interested institutions.
展开▼
