Resilience intruder alarm system using IoT approach / Wan Amir Ashraf Wan Zakaria

Criminal activities like house breaking these days are increasing in Malaysia which needs to be concerned about. One way to avoid intruders from breaking our house is by installing an alarm system as a kind of deterrent. This project focuses on helping homeowners to detect intruders from infiltrating their house area using Internet of Things (IoT) approach. Internet of Things (IoT) has been applied to various applications due to its low cost and easy to build as well as improved efficiency. The scope of this project is mainly to design an intruder alarm system that can notify homeowners via smartphone through email and SMS using GSM module as well as send data to the database when motion detected by unknown intruders within a coverage area. In this project, a microcomputer named raspberry pi model 3B+ and double pyroelectric infrared (PIR) which act as sensors were used to sense motion and detect the presence of human being or intruder. In this contact, a buzzer which attached to the raspberry pi was triggered whenever motion was notified as well as captured photo from the attached webcam that was installed for evidence. This alarm was powered by the electricity (AC) but in order to make it resilience or counter difficulties such as power failure, an additional portable battery (Powerbank Veger VP-1056) was mounted as a backup system to make the alarm remain active during power blackout. Certain constraints about the developed system and improvement of resilient intruder alarm system using IoT as compared to the conventional alarm system were highlighted. Also mentioned in this project is to add features on the webcam using Deep Learning and OpenCV that can identify and differentiate the types of detected intruders either human beings, animals or other moving objects. Ideally, such matters can be proposed for future development of this project.