Smart Cities Mobility Monitoring through Automatic License Plate Recognition and Vehicle Discrimination

This work deals with vehicular monitoring within smart cities through Automatic License Plate Recognition (ALPR) techniques and vehicle discrimination object detection (YOLO), in order to obtain timely statistical data. The combined use of the two techniques allows to obtain much more refined data than a simple vehicle counter. Moreover, the collected data undergoes a process of anonymization in accordance with the European regulations for the protection of personal data (GDPR). The use of convolutional neural networks (CNN) made it possible to obtain vehicle tracking statistics, returning daily, weekly, monthly and yearly habits with the ultimate goal of allowing a monitoring and control of the city traffic conditions. The results obtained showed a high accuracy in the classification of vehicles and a wide range of statistics concerning the occurrences of each vehicle within the area of interest.