PosturaSenseAI: AI-Powered Posture Assessment

Correct posture is essential to maintain musculoskeletal health, but people tend to fall prey to poor postural practices that culminate in long-term complications. This work proposes PosturaSenseAI, a sophisticated posture detection and correction system, which not only detects postural aberrations but also offers real-time corrective feedback and long-term monitoring. OpenCV is utilised for preprocessing and a Convolutional Neural Network (CNN) for joint detection, and the system computes important angles through trigonometry (cosine function) to evaluate posture. Real-time suggestions are made through RULA/REBA scoring, providing accurate corrective suggestions like “straighten your left shoulder.” Furthermore, a repository of posture allows users to monitor their improvement over time, representing improvement through graphed RULA/REBA scores. Experimental tests emphasise the precision and efficiency of the system in practical applications, rendering a useful tool for posture correction. The present work presents a detailed examination of methodologies applied, experimental findings, and the possible influence of our model on ergonomics and health monitoring.