Abstract

In our current digital phase, ensuring security and safety is paramount, especially when it comes to protecting our homes. Traditional methods, like relying on keys, pose vulnerabilities that can result in oversights and potential security ruptures. A robust home security system is necessary for addressing these concerns. Historically, people have secured their homes with keys, but the risk of theft increases when residents forget to lock their doors. To tackle this challenge, the research at hand leverages Deep Learning and Internet of Things (IoT) technology to elevate home security. The proposed system introduces a door lock mechanism based on facial recognition to ensure that only authorized individuals, such as friends and family, can access the premises, thereby deterring intruders. This forward-thinking solution goes beyond homes, extending its application to workplaces and campuses. Utilizing facial recognition as a seamless means of unlocking doors eliminates the need for physical effort. The system incorporates biometric and two-factor authentication, enhancing security with the integration of OpenCV. This concept combines biometric matching, human face recognition, and Twilio service-powered One-Time Password (OTP) transmission. It is powered by a Raspberry Pi 4 microprocessor. Although face recognition-based door locking systems have been around for a while, this research stands out since it incorporates extra security elements while still being reasonably priced. Consequently, it presents a comprehensive solution for heightened security in various settings.