Head Counting in Crowded Scenes Using YOLOv10: A Deep Learning Approach

Raghavendra V Vadavadagi, Sukul E N, Ankush Marlinganavvar, Anurag Hurkadli, Kunal Bhoomraddi, Uday Kulkarni

2025

Abstract

Crowd counting plays a critical role in various applications, including public safety, event management, and resource planning, by accurately estimating the number of individuals in crowded environments. This study explores the use of the advanced YOLOv10 object detection framework for counting people in such settings. By leveraging image augmentation techniques, the dataset was enhanced to improve the model’s robustness and ability to handle challenges like occlusion, overlapping objects, and varying lighting conditions. The YOLOv10 model demonstrated strong performance, achieving 49% validation accuracy at an IoU of 0.5 and 39% accuracy across IoU thresholds ranging from 0.5 to 0.9. These results underscore the model’s effectiveness in real-world crowd detection, even under complex circumstances. The model’s real-time detection capability makes it highly suitable for surveillance systems and other applications with limited computational resources. By integrating YOLOv10 into such systems, this work offers a scalable, efficient solution for accurate crowd counting, supporting safer and more efficient management of crowded scenarios. The model’s potential for further improvements, such as hyperparameter tuning, extended training, and data augmentation, promises even greater performance and scalability in future deployments.

Paper Citation

in Harvard Style

Vadavadagi R., E N S., Marlinganavvar A., Hurkadli A., Bhoomraddi K. and Kulkarni U. (2025). Head Counting in Crowded Scenes Using YOLOv10: A Deep Learning Approach. In Proceedings of the 3rd International Conference on Futuristic Technology - Volume 3: INCOFT; ISBN 978-989-758-763-4, SciTePress, pages 611-618. DOI: 10.5220/0013633400004664

in Bibtex Style

@conference{incoft25,
author={Raghavendra Vadavadagi and Sukul E N and Ankush Marlinganavvar and Anurag Hurkadli and Kunal Bhoomraddi and Uday Kulkarni},
title={Head Counting in Crowded Scenes Using YOLOv10: A Deep Learning Approach},
booktitle={Proceedings of the 3rd International Conference on Futuristic Technology - Volume 3: INCOFT},
year={2025},
pages={611-618},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013633400004664},
isbn={978-989-758-763-4},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 3rd International Conference on Futuristic Technology - Volume 3: INCOFT
TI - Head Counting in Crowded Scenes Using YOLOv10: A Deep Learning Approach
SN - 978-989-758-763-4
AU - Vadavadagi R.
AU - E N S.
AU - Marlinganavvar A.
AU - Hurkadli A.
AU - Bhoomraddi K.
AU - Kulkarni U.
PY - 2025
SP - 611
EP - 618
DO - 10.5220/0013633400004664
PB - SciTePress