Advanced Zebra Crosswalk Detection Using Deep Learning Techniques for Smart Transportation Systems
Md. Muktar Ali1, Tariqul Islam2
1Md. Muktar Ali, Department of Computer Science and Engineering, Daffodil International University, Dhaka, Bangladesh.
2Tariqul Islam, Department of Computer Science and Engineering, Daffodil International University, Dhaka, Bangladesh.
Manuscript received on 05 January 2025 | First Revised Manuscript received on 10 January 2025 | Second Revised Manuscript received on 17 January 2025 | Manuscript Accepted on 15 February 2025 | Manuscript published on 28 February 2025 | PP: 27-31 | Volume-13 Issue-3, February 2025 | Retrieval Number: 100.1/ijese.B104214020125 | DOI: 10.35940/ijese.B1042.13030225
Open Access | Editorial and Publishing Policies | Cite | Zenodo | OJS | SSRN | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: The growing prevalence of road traffic accidents poses significant challenges to vulnerable populations, particularly visually impaired individuals. To enhance safety and accessibility within smart transportation systems, this study presents an advanced detection system for zebra crosswalks, vehicles, and pedestrians utilizing deep learning techniques. The system leverages the Single Shot MultiBox Detector (SSD) model with Transfer Learning for rapid convergence and high accuracy. The dataset, derived from real-world street scenarios, includes nine object classes, enabling the system to provide real-time detection and monitoring. Experimental results demonstrate high precision and recall, underscoring the system’s potential to improve road safety and assist traffic management. Future developments will focus on integrating this system into portable devices for broader application.
Keywords: Single Shot MultiBox Detector (SSD), Traffic Monitoring, Visually Impaired Assistance, Object Detection, Pedestrian Detection.
Scope of the Article: Image Analysis and Processing