Vehicle Classification in Traffic Surveillance System using YOLOv3 Model
Abhishek Shekade, Rituja Mahale, Rushikesh Shetage, Ajit Singh, Prashant Gadakh · 2020 International Conference on Electronics and Sustainable Communication Systems (ICESC) · 2020
Vehicle detection and classification plays a vital role within the space of the traffic management system. There's an outsized area for the development during this system as associated with accuracy and exactness. Because of increasing traffic within the advanced occasions, it's basic to arrange a framework winning to keep up a record of vehicles going through a path or a street. Spontaneous identification of vehicle data has been broadly used in the vehicle identification and classification system. Applications of the system developed are often useful in the traffic signal controller, vehicle lane departure warning system. The techniques goal is to provide appropriate data about traveling vehicles with the exact count. The convolutional neural network technique models based on YOLOv3 is used. The input is given in the form of video and pre-processing is finished and also the output is gained i.e. the count of vehicles, classification of vehicles supported its sort and total variety of vehicle motion at a specific time.