Face Recognition Based Attendance System

Face Recognition Based Attendance System

Author Name : Er. Triveni Palorkar, Rajvi Patil, Aachal Meshram, Aditya Bhaisare

ABSTRACT This paper presents a Smart Attendance System utilizing Convolutional Neural Networks (CNN) to automate traditional attendance tracking methods. By employing facial recognition technology, the system aims to eliminate the errors and time consumption associated with manual attendance marking. Through the use of deep learning models and OpenCV libraries, student faces are detected, encoded into facial signature vectors, and matched against an enrolled database to automatically log attendance. The system's modular architecture facilitates seamless interaction between components, following an iterative agile methodology for continuous improvement. Extensive testing demonstrates over 90% accuracy for a class size of 50 students, with real-time processing completed in under 5 seconds. Optimizations strike a balance between performance and computational costs, while analytics offer valuable insights into attendance patterns. Overall, this research showcases the application of advanced deep learning and computer vision techniques to address real- world attendance tracking challenges, offering a robust and efficient solution for educational institutions and workplaces.