부산대학교 도서관

In this era, most of mankind’s activities are carried out on top of a desk, but they rarely bother to sit with the right posture and this can lead to problems like back pain. In 2013, the number of working days lost due to sickness absence in UK is 131 million days, and 23.66% of this number is contributed by back pain and neck pain victims [1]. In this paper, a preliminary study of posture recognition system has been developed, to rectify the user’s sitting posture by alerting him/her. A proper and improper sitting posture might look quite similar to each other in the eye of sensors, especially different heights and genders cause difficulties in detection. Hence, a preliminary posture recognition model that specifically tackles the recognition between a proper and an improper sitting posture has been developed. In this paper, we use Kinect sensor for the sitting postures detection, and then feed the postures data to the posture recognition models such as Support Vector Machine (SVM) and Artificial Neural Network (ANN), for training purpose. We compared these two models and found that SVM with linear kernel has the highest accuracy.