Embedded Human Detection System for Home Security

This paper presents the development of an embedded system prototype that performs the home security monitoring, through image processing and classification algorithms to detect human form. If human presence is detected, the system will send an alert message to the user. The embedded system is implemented on a Raspberry Pi 3 B, supported by a Pyroelectric Infrared Radial (PIR) motion sensor and a Raspberry Pi Camera V2. The algorithms are implemented in C language and were designed to take advantage of the hardware resources of the platform, through High Performance Computing (HPC) techniques. The selected classifier is a multilayer perceptron. This classifier obtained an accuracy of 96%, approximately.