Sitting posture recognition for computer users using smartphones and a web camera

This study developed models that automatically recognize proper / improper sitting postures using accelerometer readings from some human spinal points (thoracic, thoracolumbar and lumbar) through small, thin, and lightweight smartphones attached at those points, and by using a web camera which detects the upper body points' location and distances (chin, manubrium and acromion process). 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 the RapidMiner's Decision Tree using the data collected from 60 participants of different gender and body frames. Decision Tree classifier demonstrated the most promising model performance with an accuracy of 89.83% and a kappa of 0.785 for the spinal posture, and an accuracy of 95.35% and a kappa of 0.907 for head and shoulder posture. Results also showed that there were relationships between body frame and sitting posture.