AI-BASED SMART DOOR FOR ENHANCED SECURITY AND ACCESS MANAGEMENT

Abstract

Security is at most concern for anyone nowadays, whether it's data security or security of their own home. With the advancement of technology and the increasing use of IOT and AI, digital door locks have become very common these days. Face recognition system is broadly used for human identification because of its capacity to measure the facial points and recognize the identity in an unobtrusive way. The application of face recognition systems can be applied to surveillance at home, workplaces, and campuses, accordingly. The problem with existing face recognition systems is that they either rely on the facial key points and landmarks or the face embeddings from Face Net for the recognition process. Deep convolution neural networks have been successfully applied to face detection recently. Despite making remarkable progress, most of the existing detection methods only localize each face using a bounding box, which cannot segment each face from the background image simultaneously. We implement an IoT-based smart door lock system using an ESP32-CAM and an Android application in this project. Most of the time in our daily life, we forget to lock our doors and later we suffer from confusion about whether we locked all doors perfectly or not. In this project, we implement a smart door lock system, by which the owner can see the visitor’s picture and then lock or unlock their doors from anywhere and at any time using the Android application. the design and implementation of an AI-enabled smart door system capable of person identification, with the additional functionality of SMS notification. In today’s context of increasing security concerns and the need for seamless access control, this technology offers an innovative and efficient solution for recognizing authorized individuals and notifying homeowners or administrators about entry events. The smart door system harnesses the power of Artificial Intelligence (AI) algorithms, particularly Convolution Neural Networks (CNNs), to process live video streams and identify persons approaching the door. The CNN model is trained on a comprehensive dataset comprising diverse images of individuals, captured under various lighting conditions, angles, and poses, ensuring robustness and accuracy in person recognition.