Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Leveraging 3D LiDAR Sensors to Enable Enhanced Urban Safety and Public Health: Pedestrian Monitoring and Abnormal Activity Detection

Nawfal Guefrachi, Jian Shi, Hakim Ghazzai, Ahmad Alsharoa

TL;DR

This work addresses the need for accurate urban pedestrian monitoring and public health insights by leveraging elevated LiDAR and IoT. It presents a three-phase framework: (1) Blender-based synthetic data generation to overcome urban data scarcity, (2) PV-RCNN-based 3D object detection to identify pedestrians and vehicles, and (3) PointNet-based binary classification of pedestrian activities (Normal vs Abnormal) using pedestrian-specific point clouds. The approach demonstrates superior detection performance over a voxel-only baseline and robust activity classification, with a irrigation of precise sensor placement (e.g., sensors mounted at $3$ m height) to maximize coverage. The results suggest practical impact for traffic management and health interventions, enabling safer urban environments through behavioral insights and proactive public health strategies.

Abstract

The integration of Light Detection and Ranging (LiDAR) and Internet of Things (IoT) technologies offers transformative opportunities for public health informatics in urban safety and pedestrian well-being. This paper proposes a novel framework utilizing these technologies for enhanced 3D object detection and activity classification in urban traffic scenarios. By employing elevated LiDAR, we obtain detailed 3D point cloud data, enabling precise pedestrian activity monitoring. To overcome urban data scarcity, we create a specialized dataset through simulated traffic environments in Blender, facilitating targeted model training. Our approach employs a modified Point Voxel-Region-based Convolutional Neural Network (PV-RCNN) for robust 3D detection and PointNet for classifying pedestrian activities, significantly benefiting urban traffic management and public health by offering insights into pedestrian behavior and promoting safer urban environments. Our dual-model approach not only enhances urban traffic management but also contributes significantly to public health by providing insights into pedestrian behavior and promoting safer urban environment.

Leveraging 3D LiDAR Sensors to Enable Enhanced Urban Safety and Public Health: Pedestrian Monitoring and Abnormal Activity Detection

TL;DR

This work addresses the need for accurate urban pedestrian monitoring and public health insights by leveraging elevated LiDAR and IoT. It presents a three-phase framework: (1) Blender-based synthetic data generation to overcome urban data scarcity, (2) PV-RCNN-based 3D object detection to identify pedestrians and vehicles, and (3) PointNet-based binary classification of pedestrian activities (Normal vs Abnormal) using pedestrian-specific point clouds. The approach demonstrates superior detection performance over a voxel-only baseline and robust activity classification, with a irrigation of precise sensor placement (e.g., sensors mounted at m height) to maximize coverage. The results suggest practical impact for traffic management and health interventions, enabling safer urban environments through behavioral insights and proactive public health strategies.

Abstract

The integration of Light Detection and Ranging (LiDAR) and Internet of Things (IoT) technologies offers transformative opportunities for public health informatics in urban safety and pedestrian well-being. This paper proposes a novel framework utilizing these technologies for enhanced 3D object detection and activity classification in urban traffic scenarios. By employing elevated LiDAR, we obtain detailed 3D point cloud data, enabling precise pedestrian activity monitoring. To overcome urban data scarcity, we create a specialized dataset through simulated traffic environments in Blender, facilitating targeted model training. Our approach employs a modified Point Voxel-Region-based Convolutional Neural Network (PV-RCNN) for robust 3D detection and PointNet for classifying pedestrian activities, significantly benefiting urban traffic management and public health by offering insights into pedestrian behavior and promoting safer urban environments. Our dual-model approach not only enhances urban traffic management but also contributes significantly to public health by providing insights into pedestrian behavior and promoting safer urban environment.
Paper Structure (13 sections, 4 figures, 5 tables)

This paper contains 13 sections, 4 figures, 5 tables.

Table of Contents

  1. Introduction
  2. Proposed Methodology
  3. Elevated LiDAR-based Pedestrian Monitoring
  4. Data Collection
  5. Collection of Raw Point Cloud Data
  6. Data Annotation
  7. 3D Object Detection
  8. Pedestrian Activity Classification
  9. Results and Discussion
  10. Dataset and Evaluation Metrics
  11. Detection Results
  12. Classification Results
  13. Conclusion

Figures (4)

  • Figure 1: Workflow of our proposed framework for pedestrian detection and activity classification.
  • Figure 2: Screenshot of a simulated scene from Blender with an elevated LiDAR.
  • Figure 3: Dataset Visualization: This screenshot highlights pedestrians. The color variations stem from different reflectivity values of the point cloud components.
  • Figure 4: Confusion matrix showcasing the classification results of pedestrian activities into 'Normal' and 'Abnormal'.