Efficient Traffic Density Prediction Using EfficientNetB3 and YOLOv8 Models

Traffic is a big burden to government as well as public. People lives in urban areas will suffer more due to traffic. Right traffic prediction is helpful to clear crowded areas. Ambulance is facing movement issue to carry patient to hospital when crowd exist due to traffic. Efficient traffic monitoring will helpful for the government as well as public to take care their travel plan. Fixed-time traffic signals do not respond to real-time situations, and urban traffic congestion causes delays, fuel waste, and air pollution. Deep learning models such as YOLOv8, EfficientNetB3 and Convolution Neural Network(CNN) used in proposed research to monitor traffic density. The accuracy results of proposed method suggest YOLOv8 is better than other models such as CNN and EfficientNetB3 and CNN to detect vehicles in traffic. YOLOv8 model improve the prediction of traffic with accuracy 97%. Vehicle clearance, waiting time reduction, and traffic flow rate are the advantages of proposed research over existing models.

Ҳали таржима қилинмаган