Smart Blue Light Pole-Based Real-Time Crowd Counting for Smart Campuses

Crowd counting on a smart campus can provide valuable insights into pedestrian behaviors and is critical to making decisions about campus designs and improvements. Conventional image or video based crowd-counting approaches have serious privacy risks, while cloud-based solutions also face challenges in providing real-time responses. This paper proposes Height-Aware Human Classifier for Crowd Counting (HAWC-CC), a real-time, cost-effective, and privacy-protected smart blue light pole-based crowd-counting framework for smart campuses, using Light Detection and Range (LiDAR) sensors and computing devices installed on the blue light poles to collect traffic data and analyze patterns in real time while protecting data privacy. HAWC-CC is built upon two novel methods. The first is an adaptive clustering approach that dynamically adjusts to point cloud structure and density for accurate and efficient clustering. The second method is a Height-Aware Human Classifier (HAWC) that projects 3D point clouds into height-augmented multiple 2D views and uses a lightweight CNN to detect humans from the projected views. HAWC-CC and its quantized version are then implemented on Nvidia Jetson and Coral TPU for real-world deployment on the blue light poles. A comprehensive 3D LiDAR dataset is collected and curated for single-person detection and crowd counting. An extensive evaluation shows that HAWC-CC significantly outperforms the representative related works (PointNet-Cc, AutoEncoder-Cc, and OC-SVM-CC) by up to 86.61% in Mean Absolute Error (MAE) and 90.44% in Mean Squared Error (MSE). Additionally, HAWC-CC is the only framework that meets the real-time requirements for LiDAR-based crowd counting, processing one LiDAR sample in 17.42 ms. HAWC demonstrates the highest robustness to limited training data, achieving a notable accuracy of 90.29% with only 0.1% of the training data. HAWC-CC outperforms SOTA RGB image-based solutions in high-density settings, achieving 97.64% accuracy.

• Publication year

  2025

• Venue

  IEEE International Conference on Distributed Computing Systems

• Publication date

  2025-07-21

• Fields of study

  Computer Science, Engineering, Environmental Science

• Identifiers
  DOI 10.1109/ICDCS63083.2025.00098
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Heterogeneous Dual-Attentional Network for WiFi and Video-Fused Multi-Modal Crowd Counting

  2024cited by this paper
• Fusion attention Mechanic Crowd counting Network Based on Transformer

  2024influential reference
• LiDAR-PTQ: Post-Training Quantization for Point Cloud 3D Object Detection

  2024cited by this paper
• Rethinking Range View Representation for LiDAR Segmentation

  2023influential reference
• Crowdotic: Transformer-based Occupancy Estimation for Hospital Waiting Rooms with Non-speech Audio and Differential Privacy

  2023cited by this paper
• People Counting with Carry-on RFID Tags

  2023cited by this paper
• Point-Syn2Real: Semi-Supervised Synthetic-to-Real Cross-Domain Learning for Object Classification in 3D Point Clouds

  2022cited by this paper
• Rev: A Video Engine for Object Re-identification at the City Scale

  2022cited by this paper
• The Caltech Fish Counting Dataset: A Benchmark for Multiple-Object Tracking and Counting

  2022cited by this paper
• MAC address randomization tolerant crowd monitoring system using Wi-Fi packets

  2021cited by this paper
• TransCrowd: weakly-supervised crowd counting with transformers

  2021cited by this paper
• Density-Net: A Density-Aware Network for 3D Object Detection

  2021influential reference
• Unveiling personnel movement in a larger indoor area with a non-overlapping multi-camera system

  2021cited by this paper
• Cross-Modal Collaborative Representation Learning and a Large-Scale RGBT Benchmark for Crowd Counting

  2020influential reference
• Attention Scaling for Crowd Counting

  2020cited by this paper
• BirdNet+: End-to-End 3D Object Detection in LiDAR Bird’s Eye View

  2020cited by this paper
• Efficient Bird Eye View Proposals for 3D Siamese Tracking

  2019influential reference
• dbscan: Fast Density-Based Clustering with R

  2019cited by this paper
• Switching Convolutional Neural Network for Crowd Counting

  2017cited by this paper
• Single-Image Crowd Counting via Multi-Column Convolutional Neural Network

  2016cited by this paper
• PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation

  2016influential reference
• Person tracking and following with 2D laser scanners

  2015cited by this paper
• Order Matters: Sequence to sequence for sets

  2015cited by this paper
• Cross-scene crowd counting via deep convolutional neural networks

  2015cited by this paper
• Autoencoder for words

  2014influential reference
• An extensive comparative study of cluster validity indices

  2013cited by this paper
• Algorithms for hierarchical clustering: an overview

  2012cited by this paper
• Are we ready for autonomous driving? The KITTI vision benchmark suite

  2012cited by this paper
• Density‐based clustering

  2011cited by this paper
• Estimating the Support of a High-Dimensional Distribution

  2001influential reference
• Support Vector Method for Novelty Detection

  1999cited by this paper
• Data clustering: a review

  1999cited by this paper
• Using occupancy grids for mobile robot perception and navigation

  1989cited by this paper
• Pedestrian planning and design

  1971cited by this paper

• No citing papers are available for this paper.