APPLICATION DEVELOPMENT of VEHICLE COUNTING and CLASSIFICATION USING YOLOV2

0Citations
Citations of this article
11Readers
Mendeley users who have this article in their library.

This article is free to access.

Abstract

In this research, we going to introduce a method named You Only Look Once v2 (YOLOv2) for vehicle detection and vehicle classification. The vehicle is going to classify is divided by 4 there are car, motorcycle, truck, and bus. The result of classified data it can provide data for Intelligent Transportation System (ITS) to engineering the traffic road to reduce the number of traffic jam in the big city. With this method we can receive higher accuracy and performance than another method proposed before. By using data consist of 21198 objects for training, from all that data there is consist of 8936 cars, 462 buses, 1681 trucks, and 10119 motorcycles. Then from all that data we can created some weight that took for 10 hours for training so that we can receive mean Average Precision (mAP) for 60.63 % mAP and higher performance detection with frame rate 20 up to 35 frame per seconds which is almost equal to real-time.

Cite

CITATION STYLE

APA

Pryandika, Y. N., & Renny Pradina, K. (2020). APPLICATION DEVELOPMENT of VEHICLE COUNTING and CLASSIFICATION USING YOLOV2. In Journal of Physics: Conference Series (Vol. 1569). IOP Publishing Ltd. https://doi.org/10.1088/1742-6596/1569/2/022003

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free