Implementación de un sistema de rastreo y encendido automático vehicular, mediante reconocimiento facial

Abstract

The considerable increase in the theft of vehicles registered in recent years in the country, through cloning methods of keys and universal controls, tests the true effectiveness of automotive securities at the time of this type of crime. Consequently, the owner thinks twice before leaving his vehicle parked on the street. To address this problem, an automatic vehicular ignition system was implemented through facial recognition, based mainly on a new generation microcomputer, which is mainly responsible for detecting and recognizing the true owner of the vehicle, making use of artificial vision through a camera. In the first instance, all the components that make up the prototype were analyzed, starting with the central module, a Raspberry pi 3 B card; With the help of the OpenCV library and Python 3, facial recognition is carried out using the LBP (Local Binary Pattern) algorithm. Additionally, a GPS module will provide the location in real time, with the help of a web application located in a laboratory of the EPN. In addition, an alert email will be sent, indicating the location where the vehicle was turned on. Finally, all the functional tests are carried out, both in software and in hardware.