Crowd Gathering Detection Method Based on Multi-Scale Feature Fusion and Convolutional Attention
Open Access
Abstract
With rapid urbanization and growing population inflows into metropolitan areas, crowd gatherings have become increasingly frequent and dense, posing significant challenges to public safety management. Although existing crowd gathering detection methods have achieved notable progress, they still face major limitations: most rely heavily on local texture or density features and lack the capacity to model contextual information, making them ineffective under severe occlusions and complex backgrounds. Additionally, fixed-scale feature extraction strategies struggle to adapt to crowd regions with varying densities and scales, and insufficient attention to densely populated areas hinders the capture of critical local features. To overcome these challenges, we propose a point-supervised framework named Multi-Scale Convolutional Attention Network (MSCANet). MSCANet adopts a context-aware architecture and integrates multi-scale feature extraction modules and convolutional attention mechanisms, enabling it to dynamically adapt to varying crowd densities while focusing on key regions. This enhances feature representation in complex scenes and improves detection performance. Extensive experiments on public datasets demonstrate that MSCANet achieves high counting accuracy and robustness, particularly in dense and occluded environments, showing strong potential for real-world deployment.Keywords
Funding Information
- Kashgar Science and Technology Plan Project (KS2024024)
This publication has 25 references indexed in Scilit:
- AutoScale: Learning to Scale for Crowd CountingInternational Journal of Computer Vision, 2022
- Global Attention Mechanism: Retain Information to Enhance Channel-Spatial InteractionsarXiv, 2021
- Cell Localization and Counting Using Direction Field MapIEEE Journal of Biomedical and Health Informatics, 2021
- A Self-Training Approach for Point-Supervised Object Detection and Counting in CrowdsIEEE Transactions on Image Processing, 2021
- Decoupled Two-Stage Crowd Counting and BeyondIEEE Transactions on Image Processing, 2021
- Locate, Size and Count: Accurately Resolving People in Dense Crowds via DetectionIEEE Transactions on Pattern Analysis and Machine Intelligence, 2020
- Counting Objects by Blockwise ClassificationIEEE Transactions on Circuits and Systems for Video Technology, 2019
- Cross-Level Parallel Network for Crowd CountingIEEE Transactions on Industrial Informatics, 2019
- Where Are the Blobs: Counting by Localization with Point Supervision2018
- Dense crowd counting from still images with convolutional neural networksJournal of Visual Communication and Image Representation, 2016