Sitting Posture Sensing And Healthy Behavior Persuation

Sitting is the most common posture in modern occupational workspace. There is strong evidence that awkward sitting posture will increase the risk of work-related musculoskeletal disorders (WMSDs), so keeping healthy postures in daily life is important for people to maintain fitness and avoid chronic diseases. However, due to the limitations of the prior work, for both traditional sensor based method and machine vision based method, continuously observing and understanding one’s sitting posture in workplace is very difficult. At the same time, there are challenges in how to promote the occupational population to be aware of awkward sitting posture problem, change awkward sitting posture and maintain a healthy sitting posture habit. Ubiquitous Computing provides ideas and methods to address these problems. Model and study user sitting posture with the help of sensors and computer, persuade the occupational population to maintain healthy sitting posture in a positive and effective way.The main contributions of this paper are:1) Discuss about the positive relationship between WMSDs and awkward sitting posture in detail, summarize the existing sitting posture detection methods as well as their advantages and disadvantages;2) Propose a reasonable sitting posture model and a depth camera based fine-grained, real-time and non-immersive sitting posture detection algorithm with a high accuracy of92%;3) Design and implement a playful, practical and effective healthy sitting posture persuasive system based on occupational therapy, perform experiments to evaluate the validity of the persuasive system.