The College of Education for Pure Sciences, Department of Computer Science at the University of Basra, discussed a master’s thesis on designing a secure system to monitor the entry and exit of cars using image processing techniques and deep learning techniques.
The message presented by the researcher (Ghaida Youssef Abbas) included
This work focuses on securing a specific work area from unauthorized intrusion by vehicles that are not authorized to enter that work area. As well as securing the entry and exit of cars and parking lots for licensed cars in an organized and easy manner and without crowding.
In this work, the license plate recognition technology (LPR) was used to identify the license plates of Iraqi cars to identify the cars that are authorized to enter the work area from other cars.
This work includes building a database that includes the information of cars licensed to enter the work area.
The work begins with taking a picture of the vehicle that intends to enter the work area through a camera connected to the Raspberry Pi and an IR sensor that senses the arrival of the vehicle.
the cars.
Determining whether the car is allowed to enter the work area uses deep learning techniques (SSD and CCN convolutional neural network).
The message was intended
The aim of this work is to discover and identify the registration plates of Iraqi cars to secure their entry and exit to a specific work area in an organized and easy way.
The goal was also to provide a successful method for recognizing numbers, letters and words written in Arabic using deep learning.
deduce the message
Since each country has its own design of car license plates, in this
Work A system is designed to identify Iraqi vehicle license plates and use them to secure a specific work area.
The work was divided into two parts, the hardware part and the software part.
The hardware part includes a server, a Raspberry Pi, an image capture camera, and a vehicle access sensor.
The software part includes all the processing and operations that take place from the moment the vehicle image is taken until the vehicle identification is obtained.
The application of this system on 500 images of Iraqi cars gave a success rate of 97%.