Design and Build a Teacher Attendance Application Using Geolocation and Face Recognition

Abstract

Technological advances have enabled the automation of various processes, including attendance recording in educational environments. This study aims to design and build an Android-based Teacher Attendance Application with Geolocation and Face Recognition technology using the Convolutional Neural Network (CNN) method. This system was developed to replace the manual attendance method that is still applied at SMKN 1 Lintau Buo, which is vulnerable to data falsification and delays in attendance recapitulation. The method used in developing this application is the Waterfall model, including needs analysis, system design, implementation, testing, and maintenance. Geolocation is used to ensure attendance is carried out within the school area, while CNN-based Face Recognition is used to automatically verify teacher identity. This system is built with the Flutter framework for mobile applications and Flask for the backend connected to a MySQL database. The test results show that this system is capable of recording attendance with a high level of accuracy and increasing the efficiency of attendance management. With this digital system, attendance data can be stored in real-time and accessed by the admin for automatic reporting. The implementation of this system is expected to improve the accuracy, efficiency, and transparency in recording teacher attendance in educational environments