Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Detecting Unauthorized Vehicles using Deep Learning for Smart Cities: A Case Study on Bangladesh

Sudipto Das Sukanto, Diponker Roy, Fahim Shakil, Nirjhar Singha, Abdullah Asik, Aniket Joarder, Mridha Md Nafis Fuad, Muhammad Ibrahim

TL;DR

This paper tackles the challenge of detecting unauthorized auto-rickshaws in Bangladeshi traffic using real-time deep learning. It builds a domain-specific dataset by manually annotating 1,331 images with two classes (auto, non-auto) and trains a YOLOv8n detector, achieving $mAP_{50}=83.447\%$ and binary precision/recall above $78\%$. The dataset and pipeline address region-specific vehicle characteristics, enabling scalable smart-city traffic surveillance in developing contexts. The authors also discuss deployment considerations and future work, including live traffic integration and edge-device optimization, to enhance practical impact.

Abstract

Modes of transportation vary across countries depending on geographical location and cultural context. In South Asian countries rickshaws are among the most common means of local transport. Based on their mode of operation, rickshaws in cities across Bangladesh can be broadly classified into non-auto (pedal-powered) and auto-rickshaws (motorized). Monitoring the movement of auto-rickshaws is necessary as traffic rules often restrict auto-rickshaws from accessing certain routes. However, existing surveillance systems make it quite difficult to monitor them due to their similarity to other vehicles, especially non-auto rickshaws whereas manual video analysis is too time-consuming. This paper presents a machine learning-based approach to automatically detect auto-rickshaws in traffic images. In this system, we used real-time object detection using the YOLOv8 model. For training purposes, we prepared a set of 1,730 annotated images that were captured under various traffic conditions. The results show that our proposed model performs well in real-time auto-rickshaw detection and offers an mAP50 of 83.447% and binary precision and recall values above 78%, demonstrating its effectiveness in handling both dense and sparse traffic scenarios. The dataset has been publicly released for further research.

Detecting Unauthorized Vehicles using Deep Learning for Smart Cities: A Case Study on Bangladesh

TL;DR

This paper tackles the challenge of detecting unauthorized auto-rickshaws in Bangladeshi traffic using real-time deep learning. It builds a domain-specific dataset by manually annotating 1,331 images with two classes (auto, non-auto) and trains a YOLOv8n detector, achieving and binary precision/recall above . The dataset and pipeline address region-specific vehicle characteristics, enabling scalable smart-city traffic surveillance in developing contexts. The authors also discuss deployment considerations and future work, including live traffic integration and edge-device optimization, to enhance practical impact.

Abstract

Modes of transportation vary across countries depending on geographical location and cultural context. In South Asian countries rickshaws are among the most common means of local transport. Based on their mode of operation, rickshaws in cities across Bangladesh can be broadly classified into non-auto (pedal-powered) and auto-rickshaws (motorized). Monitoring the movement of auto-rickshaws is necessary as traffic rules often restrict auto-rickshaws from accessing certain routes. However, existing surveillance systems make it quite difficult to monitor them due to their similarity to other vehicles, especially non-auto rickshaws whereas manual video analysis is too time-consuming. This paper presents a machine learning-based approach to automatically detect auto-rickshaws in traffic images. In this system, we used real-time object detection using the YOLOv8 model. For training purposes, we prepared a set of 1,730 annotated images that were captured under various traffic conditions. The results show that our proposed model performs well in real-time auto-rickshaw detection and offers an mAP50 of 83.447% and binary precision and recall values above 78%, demonstrating its effectiveness in handling both dense and sparse traffic scenarios. The dataset has been publicly released for further research.

Paper Structure

This paper contains 14 sections, 5 figures, 2 tables, 2 algorithms.

Table of Contents

  1. Introduction
  2. Related Work
  3. Methods
  4. Data Collection and Preprocessing
  5. Data Collection
  6. Data Preprocessing
  7. Experimental Setup and Evaluation
  8. Hardware Configuration:
  9. Dataset Information:
  10. Training and Hyperparameter Tuning
  11. Detection and Inference Algorithm
  12. Validation and Qualitative Assessment
  13. Results and Discussion
  14. Conclusion

Figures (5)

  • Figure 1: Sample images without bounding box
  • Figure 2: Sample images with bounding box
  • Figure 3: Normalized confusion matrix
  • Figure 4: Performance curves of the model
  • Figure 5: Sample images of results