A bimodal biometrie student attendance system

A lot of attempts have been made to use biometrics in class attendance systems. Most of the implemented biometrie attendance systems are unimodal. Unimodal biometrie systems may be spoofed easily, leading to a reduction in recognition accuracy. This paper explores the use of bimodal biometrics to improve the recognition accuracy of automated student attendance systems. The system uses the face and fingerprint to take students' attendance. The students' faces were captured using webcam and preprocessed by converting the color images to grey scale images. The grey scale images were then normalized to reduce noise. Principal Component Analysis (PCA) algorithm was used for facial feature extraction while Support Vector Machine (SVM) was used for classification. Fingerprints were captured using a fingerprint reader. A thinning algorithm digitized and extracted the minutiae from the scanned fingerprints. The logical technique (OR) was used to fuse the two biometric data at the decision level. The fingerprint templates and facial images of each user were stored along with their particulars in a database. The implemented system had a minimum recognition accuracy of 87.83%.