A real-time embedded system for human gait analysis

The gait analysis plays an important role in different kinds of applications such as the disease symptoms, health condition, medical analysis, kinesiology and so forth. In this paper, we use one single video camera to extract the human body and establish the human body model by the distance map measured between the gravity center and human body contour. The obtained human body model includes seven parts such as head, upper body, lower body, left arm, right arm, left leg and right leg. To analyze the walking behaviour, we extract the angle and pace information of different body parts. Besides, we implement our proposed algorithm into an embedded system that can perform real-time pace analysis. The pace distance, walking speed, bending angle of each body parts and hand waving angle can be obtained in real-time at 30fps.