Sistem Keamanan Penghuni Apartemen Menggunakan Computer Vision Dan Raspberry Pi

Abstract

This research aims to design and implement a security system for apartments that
combines computer vision technology with notifications via the Telegram
application. This system is based on the Raspberry Pi 3 Model B+ to detect the face
of a person in front of the door via a connected camera, and sends notifications to
residents via Telegram messages. Thus, this system effectively and automatically
increases the level of security and provides residents with real-time access to
information about the ongoing security situation in their apartments. This research
explains the technical implementation of this system and its potential to increase the
sense of security of apartment residents. Test results for familiar faces have an
accuracy rate of 96% and for unfamiliar faces they have an accuracy rate of 96%.