Integration of improved YOLOv5 for face mask detector and auto-labeling to generate dataset for fighting against COVID-19

Abstract

One of the most effective deterrent methods is using face masks to prevent the spread of the virus during the COVID-19 pandemic. Deep learning face mask detection networks have been implemented into COVID-19 monitoring systems to provide effective supervision for public areas. However, previous works have limitations: the challenge of real-time performance (i.e., fast inference and low accuracy) and training datasets. The current study aims to propose a comprehensive solution by creating a new face mask dataset and improving the YOLOv5 baseline to balance accuracy and detection time. Particularly, we improve YOLOv5 by adding coordinate attention (CA) module into the baseline backbone following two different schemes, namely YOLOv5s-CA and YOLOV5s-C3CA. In detail, we train three models with a Kaggle dataset of 853 images consisting of three categories: without a mask “NM," with mask “M,” and incorrectly worn mask “IWM” classes. The experimental results show that our modified YOLOv5 with CA module achieves the highest accuracy mAP@0.5 of 93.9% compared with 87% of baseline and detection time per image of 8.0 ms (125 FPS). In addition, we build an integrated system of improved YOLOv5-CA and auto-labeling module to create a new face mask dataset of 7110 images with more than 3500 labels for three categories from YouTube videos. Our proposed YOLOv5-CA and the state-of-the-art detection models (i.e., YOLOX, YOLOv6, and YOLOv7) are trained on our 7110 images dataset. In our dataset, the YOLOv5-CA performance enhances with mAP@0.5 of 96.8%. The results indicate the enhancement of the improved YOLOv5-CA model compared with several state-of-the-art works.

Improved YOLOv5 Based Deep Learning System for Jellyfish Detection

“…4, including three main steps: optional selections (model, keyword, data file), trained model detection (YOLOv5 detector), and auto-labeling. It is noted that the YOLOv5 jellyfish detector source code is customized based on Ultralytics [15], while auto-labeling module and other parts are developed by our previous work [20]. As can be seen on the optional selections, there are several functions to select the inserting YOLOv5 jellyfish detector, including selection model, data collection, and the selection of a single file or multiple files.…”

“…Our model-assisted annotation application is developed with Python language as the back-end and PyQT5 as the front-end GUI. We provide the demo video of our model-assisted annotation application in a GitHub repository [56] together with training experiment results, model weights, and testing image results in a GitHub repository [57].…”

“…[56] T. N. Pham, V. H. Nguyen, J. H. Huh [Online]. Available: https://github.com/MiuMiao93/Integration-system-ofimproved-YOLOv5-face-mask-detection-and-auto-labelmodule [57] [Online]. Available: https://github.com/MiuMiao93/YOLO-based-deeplearning-system-for-jellyfish-detection/tree/main He is the Book Author of "Smart Grid Test Bed Using OPNET and Power Line Communication" pp.1-425, IGI Global, USA, 2017.…”

“…As a result, instead of following classical manual annotation methods, we introduce a model-assisted annotation method with auto-labeling, which we published in [20], to create our jellyfish dataset to save time and labor of annotation task. Therefore, the model deployed into the auto-labeling module requires fast detection time with an accurate detection rate without missing, over-detection, and better localization of bounding boxes around jellyfish.…”

Face Mask Detection using MobileNetV2 and OpenCV

“…Jayaswal and Dixit [7] proposed a face detection system using single Shot Multiboxdetector as a face detector model and a deep Inception V3 architecture (SSDIV3) to extract the pertinent features of images and discriminate them in mask and without masks labels. Pham et al, [8] proposed an improvised YOLOv5s-CA face mask detection model for categorizing between people wearing or not wearing masks. Thus it can be seen works in the literature have used deep learning models as they can work on large data effectively.…”