An Efficient and Cost Effective Performance of Autonomous Attendance System Using Convolution Neural Networks | Chapter 1 | Research and Developments in Engineering Research Vol. 4

 This research work is proposed to developing a less turmoil, cost effective and robust exercise of autonomous attendance administration system. To watch carefully students' daily occupancy, marking attendance is a low practice for all educational organizations at all levels. There was once a opportunity when attendance was tracked manually. Although they are behind and labor-intensive for a large number of pupils, these approaches are exact and expel the possibility of deceptive enrollment. To address the disadvantages of manual systems, autonomous schemes based on high frequency recognition leafing through, fingerprint scanning, face acknowledgment, and iris scanning are being grown. Each technique has benefits and disadvantages. However, the majority of these systems are restricted by the necessity for individual human engagement all along the recording of attendance. To close the gaps in the current human and independent attendance management structures, we developed a healthy and efficient attendance recording order using a distinct group shot to detect face labeling and recognition algorithms. A extreme-definition camcorder mounted in a stable district records a series of countenances for each scholar seated in a classroom. The faces are therefore extracted from the group chance using a protocol, and they are identified using a convolutional interconnected system that is familiar to students from a face table. We evaluated our technique utilizing several datasets and group pictures. Our research shows that the submitted framework performs better in agreements of effectiveness, utility, and implementation than the current attendance pursuing methods. The suggested resolution is a self-contained attendance method that necessitates little contact middle from two points humans and machines, making integration into a smart school room simple.

Author(s) Details:

S. Purushothaman,
Department of Electronics and Communication Engineering, V.S.B Engineering College, Karur, Tamil Nadu, India.

Please see the link here: https://stm.bookpi.org/RADER-V4/article/view/10738