Diseño de un sistema embebido para el control de ingreso y salida de vehículos a través de internet, en el acceso principal de la ESPOCH.

Abstract

The embedded real-time system was developed by using a reduced board computer Raspberry Pi3 which is an equipment with a camera placed in strategic points of the mechanic arms in the main entrance and exit of the Escuela Superior Politécnica de Chimborazo, this equipment captures images of the vehicles that enter and exit the campus and the information is extracted through the implementation of an algorithm of segmentation written in Python programming language and the collaboration of the artificial vision bookstores offered by OpenCV, processing techniques were applied to extract the vehicle plate from the location scenery. Then, an Optical Character Recognition (OCR) algorithm also known as K-Nearest Neighbours (KNN) was applied, which after a training phase identifies letters and numbers in the automobile plates, the information is storage in the entrance data base and deleted when exiting the site. This information is filed in a database anchored in the network, so the vehicle owner could know the time its vehicle stayed on campus by accessing to any devices with web service. The implementation of vehicles registration presents an alternative for the conventional vehicle entry systems and the advantage is that it does not require the use of device or additional sensors in the vehicle to recognize it; besides, it provides information in real time of all the data acquired on the web. This system opens the possibility to develop new applications which work according to the operation.