IMPLEMENTATION OF MOBILENETV2 AND FRAME DIFFERENCE FOR VEHICLE SPEED DETERMINATION

Authors

  • Ridho Tadjudin ITENAS Bandung
  • Dewi Rosmala Institut Teknologi Nasional

DOI:

https://doi.org/10.33197/jitter.vol7.iss3.2021.541

Keywords:

data video, frame difference, mobilenetv2, car, smart city

Abstract

Camera devices on the road can be used to detect problems such as violations of speed limits which can be fatal, lead to accidents that can lead to serious injury or death. This research focuses on a system that can calculate vehicle speed using the Frame Difference method from the video data that has been extracted first into a series of images and the object segmentation process is carried out using the MobileNetV2. Frame Difference is used to calculate the displacement value of the car object in each series of images. The displacement value is then used to determine the displacement speed of the car in a series of images. From the results obtained, the system has an accuracy of 98% for car segmented image, 74.81% accuracy for determining the reference point, and in determining the speed of the car, an accuracy of 83.84% for the right reference point and 79.81% for the left reference point is obtained.

Downloads

Download data is not yet available.

Published

2021-08-13

How to Cite

[1]
R. Tadjudin and D. Rosmala, “IMPLEMENTATION OF MOBILENETV2 AND FRAME DIFFERENCE FOR VEHICLE SPEED DETERMINATION”, jitter, vol. 7, no. 3, pp. 193–204, Aug. 2021.

Issue

Section

Articles