Akademik Veri Yönetim Sistemi
Çimtay Y. (Yürütücü), Nur Yılmaz G.
Yükseköğretim Kurumları Destekli Proje, 2022 - 2022
"Today, the increasing number of vehicles in the traffic give rise to some problems related to safety. Though many precautions are taken by the governments, still many drivers don’t obey the rules and it causes many traffic accidents which result in death or injuries of many people. One of the most dangerous events in the traffic is the vehicles which wave through the traffic. It can be defined as the zigzag movements of vehicles among the other vehicles in the traffic, generally with high speed. Although, there are several companies which develop smart solutions to traffic related problems, automatic detection and reporting of the waving through traffic event has not been solved yet. It is possible to hear and see many traffic accidents due to those kinds of events. Those kinds of behaviors are generally reported by the other drivers or detected by secret polices who follow the traffic. Today it is possible to see those kinds of drivers almost every day in Turkey. Therefore the automatic detection and reporting of waving through traffic event is required. In this project, we propose a system which can detect the waiving event and read the plate of the car and report the plate to the center automatically. An image/video processing system will be developed on the simulated traffic data. This data will be produced by using car driving games. The system is considered to be integrated to the security cameras which are already being used by General Directorate of Security for controlling the roads. The developed system will be adopted to real traffic data and fine-tuned. In this project, the state-of-the art object tracking and plate recognition methods will be used and moreover, the deep learning models like Convolutional Neural Network (CNN) and Long Short Term Memory (LSTM) and the hybrid combination of them will be adopted for the detection of waving through traffic events."