Automatic Attendance System through Video Monitoring

Abstract

In today's rapidly changing technological environment, people run after the automation for precious time saving, using reliable and effective systems. In general, marking attendance in a college, school or work environment, at regular intervals during the day requires a lot of time consumption, as well as proxy, to solve these limitations, everyone looking for automated system using bio-metrics. Bio-metrics are of different methods like finger print, iris, facial recognition etc, out of these; we follow facial features recognition system for marking attendance in class or college where targets at certain distance. This present paper proposes the methodology for facial recognition through video monitoring. For facial recognition many algorithms are available but for good resolution we follow the Kanade-Lucas-Tomasi (KLT) system. The proposed system uses the database for comparing the Principle Component Analysis(PCA) and Linear Discriminant Analysis (LDA). The video output, which is collected from the web camera is verified through data base using KLT algorithm and attendance is marked using MATLAB and the data is stored in excel data sheet.