A rehabilitation planning based on kinect somatosensory recognition and cloud computing

Abstract

In order to solve the physical rehabilitation of stroke patients with apoplectic hemiplegia, an intelligent rehabilitation supporting system based on Kinect device is designed. The system uses Kinect device to capture the user’s limb joints point information. DTW algorithm is combined with RANSAC algorithm to match the action in standard action library, thus the system can obtain the evaluation results. Then the patients’ rehabilitation data is sent to the doctors with new treatment options intelligently. As a result, the experimental results show that this physical rehabilitation method is more convenient and efficient than the traditional training method.