Citation

Khow, Zu Jun and Goh, Michael Kah Ong and Tee, Connie and Law, Check Yee (2022) A YOVO5 Based Real-time Helmet and Mask Detection System. Journal of Logistics, Informatics and Service Science, 9 (3). pp. 97-111. ISSN 2409-2665

Full text not available from this repository.

Abstract

The COVID-19 pandemic has brought unimaginable damage to the globe; It had brought people we love away and forced the government to lock down the cities to prevent the infection of COVID-19 from spreading. This stopped various industries from working, especially the engineering and construction industries. Fortunately, the effectiveness of the COVID-19 vaccine had enabled the industries to resume their operation back to normal. However, the mask now is an essential equipment to be worn by all workers while on site, they also required to wear helmet for safety reasons. Therefore, the aim of research is to detect the helmets and masks worn by workers, if there the workers were found not for not wearing the mask properly an alert will be triggered. Five classes were determined namely 'Head,’ 'Helmet,’ 'Incorrect Mask,’ 'No Wearing Mask', and 'Wearing Mask'. A total of 1711 images of construction workers scenes were collected, and augmentation was applied on these images to generate 4733 images. All images were annotated corresponding to the defined classes. The experiments have split the training and validation dataset into a ratio of 9:1. The result obtain a 65.235% Mean Average Precision (mAP) as a result.

Item Type:	Article
Uncontrolled Keywords:	Deep learning, object detection, you only look once (YOLO), mask detection, helmet detection
Subjects:	Q Science > Q Science (General) > Q300-390 Cybernetics
Divisions:	Faculty of Information Science and Technology (FIST)
Depositing User:	Ms Nurul Iqtiani Ahmad
Date Deposited:	31 Oct 2022 02:34
Last Modified:	31 Oct 2022 02:34
URII:	http://shdl.mmu.edu.my/id/eprint/10492

Downloads

Edit (login required)