Face recognition-based attendance system with anti-spoofing, system alert, and email automation

Abstract

The subject matter of the article is the design of an attendance system based on face recognition with anti-spoofing, system alarm, and Email Automation to improve accuracy and efficiency, highlighting its potential to revolutionize traditional attendance tracking methods. The administration of attendance might be a tremendous load on the authority if it is done manually. Therefore, the goal of this study is to design a reliable and efficient attendance system that can replace traditional manual approaches while also detecting and preventing spoofing attempts. Without the manual approach, attendance may be collected using many kinds of technologies, including biometric systems, radiofrequency card systems, and facial recognition systems. The face recognition attendance system stands out among the rest as a great alternative to the traditional attendance system used in offices and classrooms. The tasks to be accomplished include selecting appropriate facial detection and recognition technologies, implementing anti-spoofing measures to prevent intruders from exploiting the system, and integrating system alarms and email automation to improve accuracy and efficiency. The methods used include selecting the Haar cascade for facial detection and the LBPH algorithm for facial recognition, using DoG filtering with Haar for anti-spoofing, and implementing a speech system alarm for detecting intruders. The result of the system is a face recognition rate of 87 % and a false positive rate of 15 %. However, since the recognition rate is not 100 %, attendance will also be informed through email automation in case someone is present but is not detected by the system. In conclusion, the designed attendance system offers an effective and efficient alternative to the traditional attendance system used in offices and classrooms, providing accurate attendance records while also preventing spoofing attempts and notifying authorities of any intruders.