Leveraging Pre-Trained Vision for Enhanced Real Time Pedestrian Behavior Prediction at Zebra Crossings

Authors

  • Pannalal Boda Research Scholar, CSE Dept, Osmania University, Hyderabad, Telangana, India. Author
  • Y. Ramadevi Professor, CSE Dept, Chaitanya Bharathi Institute of Technology, Osmania University, Hyderabad, Telangana, India. Author
  • M Bhavsingh Associate Professor, Department of Computer Science and Engineering, Ashoka Women’s Engineering College Kurnool, Andhra Pradesh, India. Author

Keywords:

Pedestrian behavior prediction, real-time safety, pre-trained vision, convolutional neural networks, zebra crossings, intelligent traffic management.

Abstract

Ensuring real-time pedestrian safety at zebra crossings requires accurate prediction of their 
behavior in dynamic traffic conditions. Existing static measures often fail to adapt to these 
complexities. This paper proposes a novel approach for Leveraging Pre-Trained Vision for 
Enhanced Real-Time Pedestrian Behavior Prediction at Zebra Crossings. Traditional methods and 
models trained from scratch necessitate vast amounts of labeled data, hindering implementation. 
We address this by utilizing pre-trained convolutional neural networks (CNNs) with established 
feature extraction capabilities. These pre-trained models are then fine-tuned on a focused dataset 
of pedestrian behavior at crossings, allowing them to specialize in this specific environment. This 
significantly reduces the data collection burden while improving prediction accuracy compared 
to traditional methods. The proposed approach is evaluated using a pedestrian behavior dataset at 
zebra crossings, with performance measured by established metrics. Our research demonstrates 
that leveraging pre-trained vision models achieves superior real-time prediction accuracy, 
ultimately enhancing pedestrian safety at zebra crossings. This approach is applicable to 
intelligent traffic management systems, enabling proactive interventions to prevent accidents. 

Downloads

Published

2023-06-30

Issue

Vol. 1 No. 2 (2023): June (2023) Issue

Section

Articles

License

Public Licensing Terms

This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0).

You are free to:

  • Share: Copy and redistribute the material in any medium or format.
  • Adapt: Remix, transform, and build upon the material for any purpose, even commercially.

Under the following terms:

  • Attribution: You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.

To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

How to Cite

Pannalal Boda, Y. Ramadevi, & M Bhavsingh. (2023). Leveraging Pre-Trained Vision for Enhanced Real Time Pedestrian Behavior Prediction at Zebra Crossings . Frontiers in Collaborative Research, 1(2), 10-21. https://www.macawpublications.com/Journals/index.php/FCR/article/view/22