A Lightweight Safety Helmet Compliance Detection via Multimodal Fusion

Jeong Hwan Ryu; Azimjon Akhtamov; Md Azher Uddin; Aziz Nasridinov

doi:10.1609/aaai.v40i48.42274

A Lightweight Safety Helmet Compliance Detection via Multimodal Fusion

Jeong Hwan Ryu
, Azimjon Akhtamov
, Md Azher Uddin^*
, Aziz Nasridinov^*

^*Corresponding author for this work

School of Mathematical & Computer Sciences

Research output: Contribution to journal › Meeting abstract › peer-review

Abstract

Ensuring proper use of personal protective equipment (PPE), especially helmets, is essential for workplace safety. Conventional object detectors often fail to distinguish whether a helmet is worn correctly, and existing approaches relying on single-model pipelines are prone to localization errors and false alarms. Moreover, most prior studies do not guarantee real-time performance. To resolve these challenges, we propose a lightweight multimodal approach that integrates a YOLO11-based object detector with a pose estimation model, achieving higher F1 scores and lower false alarm rates while maintaining real-time performance.

Original language	English
Pages (from-to)	41373-41374
Number of pages	2
Journal	Proceedings of the AAAI Conference on Artificial Intelligence
Volume	40
Issue number	48
DOIs	https://doi.org/10.1609/aaai.v40i48.42274
Publication status	Published - 14 Mar 2026
Event	40th Annual AAAI Conference on Artificial Intelligence 2026 - Singapore, Singapore Duration: 20 Jan 2026 → 27 Jan 2026 https://aaai.org/conference/aaai/aaai-26/

ASJC Scopus subject areas

Artificial Intelligence

Access to Document

10.1609/aaai.v40i48.42274

Cite this

@article{ae2b826178064e538262a1787eac70bb,

title = "A Lightweight Safety Helmet Compliance Detection via Multimodal Fusion",

abstract = "Ensuring proper use of personal protective equipment (PPE), especially helmets, is essential for workplace safety. Conventional object detectors often fail to distinguish whether a helmet is worn correctly, and existing approaches relying on single-model pipelines are prone to localization errors and false alarms. Moreover, most prior studies do not guarantee real-time performance. To resolve these challenges, we propose a lightweight multimodal approach that integrates a YOLO11-based object detector with a pose estimation model, achieving higher F1 scores and lower false alarm rates while maintaining real-time performance.",

author = "Ryu, \{Jeong Hwan\} and Azimjon Akhtamov and Uddin, \{Md Azher\} and Aziz Nasridinov",

note = "Publisher Copyright: {\textcopyright} 2026, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved.; 40th Annual AAAI Conference on Artificial Intelligence 2026, AAAI 2026 ; Conference date: 20-01-2026 Through 27-01-2026",

year = "2026",

month = mar,

day = "14",

doi = "10.1609/aaai.v40i48.42274",

language = "English",

volume = "40",

pages = "41373--41374",

journal = "Proceedings of the AAAI Conference on Artificial Intelligence",

issn = "2159-5399",

publisher = "Association for the Advancement of Artificial Intelligence",

number = "48",

url = "https://aaai.org/conference/aaai/aaai-26/",

}

TY - JOUR

T1 - A Lightweight Safety Helmet Compliance Detection via Multimodal Fusion

AU - Ryu, Jeong Hwan

AU - Akhtamov, Azimjon

AU - Uddin, Md Azher

AU - Nasridinov, Aziz

PY - 2026/3/14

Y1 - 2026/3/14

N2 - Ensuring proper use of personal protective equipment (PPE), especially helmets, is essential for workplace safety. Conventional object detectors often fail to distinguish whether a helmet is worn correctly, and existing approaches relying on single-model pipelines are prone to localization errors and false alarms. Moreover, most prior studies do not guarantee real-time performance. To resolve these challenges, we propose a lightweight multimodal approach that integrates a YOLO11-based object detector with a pose estimation model, achieving higher F1 scores and lower false alarm rates while maintaining real-time performance.

AB - Ensuring proper use of personal protective equipment (PPE), especially helmets, is essential for workplace safety. Conventional object detectors often fail to distinguish whether a helmet is worn correctly, and existing approaches relying on single-model pipelines are prone to localization errors and false alarms. Moreover, most prior studies do not guarantee real-time performance. To resolve these challenges, we propose a lightweight multimodal approach that integrates a YOLO11-based object detector with a pose estimation model, achieving higher F1 scores and lower false alarm rates while maintaining real-time performance.

UR - https://www.scopus.com/pages/publications/105034260582

U2 - 10.1609/aaai.v40i48.42274

DO - 10.1609/aaai.v40i48.42274

M3 - Meeting abstract

AN - SCOPUS:105034260582

SN - 2159-5399

VL - 40

SP - 41373

EP - 41374

JO - Proceedings of the AAAI Conference on Artificial Intelligence

JF - Proceedings of the AAAI Conference on Artificial Intelligence

IS - 48

T2 - 40th Annual AAAI Conference on Artificial Intelligence 2026

Y2 - 20 January 2026 through 27 January 2026

ER -

A Lightweight Safety Helmet Compliance Detection via Multimodal Fusion

Abstract

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this