Real-time human sitting posture detection using mobile devices

Abstract

This study developed models to detect proper/ improper sitting postures using gyroscope readings from some human spinal points (thoracic, thoraco-lumbar and lumbar) through mobile devices attached at those points. It also established relationships of human body frames and proper sitting posture. The models were developed by training some well-known classifiers such as KNN, SVM, MLP, and Decision Tree using the data collected from 49 students of different body frames. Decision Tree classifier demonstrated the most promising model performance with an accuracy of 96.13% and a kappa of 0.921. Results also showed that there were relationships between body frame and posture.