AUTOMATIC EXAM ATTENDANCE SYSTEM BASED ON ILLUMINATION INVARIANT FACE RECOGNITION

Abstract

Security systems are used in several ways. In the developing technology, the computer systems are employed to solve different kinds of problems whereas they can be also used for security purposes. Application of the computer-aided security systems are one of the applicable technologies today especially in the crowded places such as entrance gates where high security measure is requested. On the other hand, automatic face recognition is useful in the applications where the recognition of the authorized people should be completed in a limited time. An application, that is the identification of the students for exam security is one of the important issues in universities where crowded exams take place. Unidentified people other than one's own examinations can be defined as problematic in terms of exam assessment. The paper proposes a new automatic class attendance system based on illumination invariant face recognition. System consists of three stages which are the face detection, facial feature extraction and classification. A known method will be employed for face detection part. For the facial feature extraction stage, non-subsampled Contourlet transform is used. The classification is done by the use of a known method which is the correlation coefficient. The system is currently under test and expected to run at acceptable recognition rates to be used in an automatic class attendance system.