An Efficient and Cost Effective Performance of Autonomous Attendance System Using Convolution Neural Networks

This research work is aimed to developing a less disruption, cost effective and robust implementation of autonomous attendance management system. To keep track of students' daily presence, marking attendance is a common practice for all educational institutions at all levels. There was once a time when attendance was tracked manually. Although they are time-consuming and labor-intensive for a large number of pupils, these approaches are exact and exclude the possibility of fraudulent enrollment. To address the drawbacks of manual systems, autonomous systems based on radio frequency recognition scanning, fingerprint scanning, face recognition, and iris scanning are being developed. Each technique has benefits and drawbacks. However, the majority of these systems are limited by the requirement for individual human engagement during the recording of attendance. To close the gaps in the current human and autonomous attendance management systems, we developed a robust and efficient attendance recording system using a single group shot to detect face identification and recognition algorithms. A high-definition camera mounted in a stable location records a series of images for each student seated in a classroom. The faces are then extracted from the group shot using a standard procedure, and they are identified using a convolutional neural network that is well-known to students from a face database. We evaluated our technique using several datasets and group pictures. Our research shows that the suggested framework performs better in terms of effectiveness, usability, and implementation than the current attendance tracking methods. The suggested solution is a self-contained attendance system that necessitates little contact between humans and machines, making integration into a smart classroom simple.