Enhancing Smart Home Security System with HOG Algorithm

Abstract

The escalating global concern over burglary incidents, particularly by the economic repercussions of the 2020 pandemic, necessitates innovative solutions to fortify residential security. Traditional locks, once presumed as the primary defence against intruders, have proven inadequate against modern burglary techniques. This work proposes the development and implementation of a "Smart Home Security System" to address the imperative need for improved home security. By implementing technologies such as the Internet of Things (IoT), artificial intelligence (AI), and mobile applications, the proposed system aims to establish multiple layers of protection against unauthorized entry. The objectives of this study revolve around the integration of various components, including smart locks employing solenoid locking mechanisms, ultrasonic sensors, video surveillance, and mobile application-based access control. The system offers diverse methods for unlocking, encompassing passcodes, face recognition utilizing the HOG (Histogram of Oriented Gradients) algorithm, mobile applications, and physical button switches. Leveraging Firebase for database management enhances system reliability and accessibility. Additionally, the Smart Home Security System incorporates important features such as real-time monitoring of the front door and image capture capabilities. Furthermore, a burglary detection system, facilitated by ultrasonic sensors, serves as a proactive measure against unauthorized access attempts. A key goal of the proposed Smart Home Security System is to offer homeowners an advanced and comprehensive security solution at a lower cost, while maintaining functionality superior to existing systems on the market. By integrating cutting-edge technology and cost-effective components, this system provides a robust and accessible defense mechanism, empowering homeowners with enhanced protection against contemporary burglary threats.