Traffic Lights Detection using Machine Learning

Authors

  • Saurav Shaurya
  • Vrushali Phaltankar
  • Tanguturi Sai Harshith
  • Reddy Sree Ranga Santosh
  • Ayanesh Chowdhury

Keywords:

real-time, traffic-lights, video-feed, blob-detection, closing-operation, MATLAB, noisy-imaging conditions

Abstract

Real-time detection of visitors lighting from video feed has numerous actual-lifestyles applications, which include self-using motors, automated train sign detection, and many others. This paper describes a way to locate pink and green round visitor’s lights from a video feed recorded by way of a digicam established on a shifting automobile. In this technique, we first set thresholds for purple and inexperienced within the RGB shade area after which become aware of blobs inside the body that meet this threshold. After blob detection, we carry out a last operation to cast off gaps. We then become aware of the centre and radius of the round blobs and draw a circle around the detected visitor’s alerts. The software has been carried out on Matlab on an Intel i5 processor and examined numerous video samples to decide detection accuracy. Further, we also examined the overall performance of the gadget under noisy imaging situations. The method meets real-time processing requirements of 15 frames in step with second.

Downloads

Download data is not yet available.

Downloads

Published

26-12-2021

Issue

Section

Articles

How to Cite

[1]
S. Shaurya, V. Phaltankar, T. S. Harshith, R. S. R. Santosh, and A. Chowdhury, “Traffic Lights Detection using Machine Learning”, IJRESM, vol. 4, no. 12, pp. 98–100, Dec. 2021, Accessed: Dec. 14, 2024. [Online]. Available: https://journal.ijresm.com/index.php/ijresm/article/view/1622

Most read articles by the same author(s)