Year: 2024
Type: conference-paper
Abstract: The well-being and productivity of employees across various industries are paramount concerns.Fatigue, often stemming from prolonged work hours and demanding conditions, significantly threatens people's health and productivity.This study focuses on fatigue detection, leveraging advanced machine learning models.Drawing from a fatigue-detection dataset obtained from Kaggle and meticulously partitioned into training, testing, and validation subsets, our research evaluates several deep learning architectures, including MobileNet, VGG16, NASNetLarge, and ResNet50.Through exhaustive experimentation and rigorous evaluation, we identify MobileNet as the top performer, demonstrating exceptional accuracy and robustness with an accuracy rate of 98.65%.Our findings highlight technological advancements in fatigue detection and emphasize prioritizing well-being.By embracing solutions like Mo-bileNet, we enhance workplace safety, productivity, and quality of life.This research guides future efforts towards a safer and more sustainable work environment across all sectors.
Institution: United International University
Language: English
Subfield: Experimental and Cognitive Psychology
Field: Psychology
Domain: Social Sciences
Open Access status: bronze
Funder: United International University
OpenAlex ID: W4411099268