Smart Traffic Light System by Using Artificial Intelligence

Zia, Sahar; Naseem, Mustafa; Malá, Ivana; Tahir, Muhammad Junaid; Mughal, T. J. A.; Mubeen, Tauseef

doi:10.26692/sujo/2018.12.00104

Cited by 4 publications

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The intelligent system needs to employ numerous modern algorithms in deep learning to automate distinct traffic control requirements. Adding on to account a large database of dedicated video is taken into examining the data, deep learning-based models are trained to determine as a queue, track stationary vehicles, and take vehicles to count in the form of lists (5).…”

Section: Introductionmentioning

confidence: 99%

Intelligent Traffic Control System Using Deep Learning

M¹,

G²,

M³

et al. 2022

ECS Trans.

View full text Add to dashboard Cite

Traffic congestion and regulating traffic in traffic signals are major issues in cities. Nowadays, in most of the cities, traffic management centers installed numerous cameras all over the roads and traffic signals. Such cameras can be effectively used for the automation of traffic signals. The objective is to develop a real time system that can automatically monitor real time traffic and make the system intelligent using artificial intelligence techniques. Specifically, Deep Convolutional Neural Networks are employed to perform the task. From statistical traffic data, it determines count, type of vehicle, average speed, distance between vehicles, etc. Based on traffic, the algorithm instructs to stop vehicle or queue or move. It can also record a wrong-way driver. Using license plate recognition, security applications such as unauthorized vehicles are identified. If there is violation of traffic rules, they are recorded with registration number. It can detect ambulances and give first preference. The proposed algorithm identifies VIP vehicles and clear traffics in automated ways. Ambulances are given priority to pass the road. The entire system have been developed using a standalone-Graphical User Interface (GUI). We have implemented successfully and the proposed framework performs satisfactorily.

show abstract

Section: Introductionmentioning

confidence: 99%