<![CDATA[This study aims to develop a web-based ergonomic sitting posture detection system to reduce postural fatigue caused by prolonged computer use. The proposed system uses a deep learning-based pose estimation method to detect body keypoints and calculate the user's posture angles. The dataset used consists of hundreds of images that have been enhanced in quality and quantity through a data augmentation process. The system then classifies sitting postures into ergonomic and non-ergonomic categories. Test results show that the system is able to achieve a high level of accuracy, with the model achieving 98% Precision, 99% Recall, 99% mAP50, and 82% mAP50-95 in detecting and classifying sitting postures. Furthermore, the web-based implementation allows for real-time monitoring. The results of this study indicate that computer vision technology has the potential to be an effective solution to increase awareness of correct sitting posture and help prevent postural fatigue, especially in academic environments.]]>
Copyrights © 2026
