IoT based smart doorbell with face recognition and remote alerts

Smart doorbells are the requirement in the houses, especially in the city for the security of the members where most of the people are outside home for work or study and only few members are inside home, sometimes only children or elderly people. Our effort is to provide a solution to this problem with a robust and cost-effective smart doorbell system that combines computer vision, Internet of Things (IoT), and cloud-based messaging to enhance residential security. When a visitor presses the doorbell button, the system captures an image using a connected webcam, performs facial recognition using a pre-trained model, rings a buzzer, and sends a real-time photo alert with the identified name via Telegram. The camera then powers down to conserve resources. The system then switches to a Blynk-controlled interface, allowing the user to remotely unlock or lock the door using two buttons on the Blynk mobile or web dashboard. The system runs on a Raspberry Pi, integrating hardware-level GPIO interaction with cloud-based APIs, offering a hybrid edge-cloud security solution.