A New Method on Mask-Wearing Detection for Natural Population Based on Improved YOLOv4

TL;DR

This paper introduces an improved YOLOv4-based method for mask-wearing detection in public places, enhancing accuracy and robustness to support automated supervision during the COVID-19 pandemic.

Contribution

It proposes specific modifications to YOLOv4, including adding a Coordinate Attention Module and adaptive anchor box clustering, to improve mask detection performance.

Findings

Improved YOLOv4 exceeds baseline by 4.06% AP.

Model achieves 64.37 FPS for real-time detection.

Enhanced robustness and accuracy in mask detection.

Abstract

Recently, the domestic COVID-19 epidemic situation is serious, but in public places, some people do not wear masks or wear masks incorrectly, which requires the relevant staff to instantly remind and supervise them to wear masks correctly. However, in the face of such an important and complicated work, it is very necessary to carry out automated mask-wearing detection in public places. This paper proposes a new mask-wearing detection method based on improved YOLOv4. Specifically, firstly, we add the Coordinate Attention Module to the backbone to coordinate feature fusion and representation. Secondly, we conduct a series of network structural improvements to enhance the model performance and robustness. Thirdly, we adaptively deploy the K-means clustering algorithm to make the nine anchor boxes more suitable for our NPMD dataset. The experiments show that the improved YOLOv4 performs better, exceeding the baseline by 4.06\% AP with a comparable speed of 64.37 FPS.