IAAS: IoT-Based Automatic Attendance System with Photo Face Recognition in Smart Campus

A lecturer’s most oppressive task is marking attendance. Manually marking the attendance of students also takes a lot of time. Although biometric systems have been used for automation in the past, it has not proven to be a very efficient method since it is much easier to fake a fingerprint today. When selecting a method for this task, accuracy and authenticity are also important considerations, something many proposed methods do not address. Nowadays, smart attendance monitoring systems such as biometric facial recognition systems are being proposed in order to eliminate manual paper-based attendance systems. There are thousands of applications for it, such as monitoring the classroom using CCTV, creating computer–human interaction, maintaining accurate attendance, and maintaining security. Students can be marked absent even when inside the classroom using this system. Auto-attendance systems rely primarily on facial recognition in contrast to other biometrics. In this study, three methods are applied: Convolution Neural Network (CNN), Viola–Jones (Haar-Cascade), and Histogram Oriented Gradients (HOG) are evaluated on the basis of face detection accuracy. Using an excel sheet, attendance information is automatically updated in the system. In the event of an absent student, a message will automatically be sent by using SMTP protocol to their parent’s phone number.