Development of object recognition system for vehicular environment

OData support
Supervisor:
Gódor Győző
Department of Networked Systems and Services

A smart city is an urban development vision to integrate multiple information and communication technology (ICT) solutions in a secure fashion to manage a city’s assets. One of these assets is the intelligent parking system, which assists for finding free parking lots reducing unnecessary searching time. For that, the purpose of this bachelor thesis is to develop a mobile application, which is able to automatically detect vehicles and distinguish them from free parking lots.

In order to implement this system, there are several steps which are including installing tools, environmental configuration, developments. I implemented both a server and an Android application using several development tools, but most important part is using cascade classifier for Haar features. This cascade classifier could be generated by training huge number of vehicle images and non-vehicle images with OpenCV’s functions.

With this trained cascade classifier, I could make an Android application which can detect the left side of vehicles in real time, and the detected vehicle could be highlighted on the screen with rectangular box. Also this application can send the position and the parking lot status to the server from the phone. By receiving data from the application, the server can modify given parking lot’s status using a connected MySQL database where the set of parking lot’s locations are stored.

Downloads

Please sign in to download the files of this thesis.