A Conceptual Model of Automated Attendance System using Image Processing

Abstract: Conventionally, student record keeping is done manually by teachers through roll calling or by passing an attendance sheet in the class. These methods are time consuming, prone to errors and proxy attendance. Moreover digital assimilation of records is tedious since teachers need to fill in the details in the database by themselves to generate reports. Consistency in manual and digital records also needs to be maintained. In the recent years, automated systems that have evolved use standard biometrics like fingerprint and iris recognition. These systems are intrusive in nature and require expensive gadgetry. Also deploying them on a large scale is detrimental to the project budget. Thus our project design makes use of facial detection and recognition algorithms like Viola Jones and PCA to identify the students in the image captured by a basic camera. This image is uploaded in the software by the professors using a simple USB cable where various combinations of algorithms detect the faces and compare it to the student image database for recognition phase. Daily attendance automatically gets updated in the database thereafter. This system solves the problem of redundancy in manual records and makes attendance keeping a convenient task.

Keywords: Automated Attendance System, Face Detection, Face Recognition, PCA (Principal Component Analysis), Viola Jones, Camera, OpenCV, PHP.