Abstract
The gait analysis always plays an important role in different kinds of applications such as the disease symptoms, health condition, medical analysis, kinesiology and so forth. In the gait analysis, we are able to obtain the information of the pace speed, pace distance, pace period, and the movement track and angle of one's feet in order to observe the physiology of a person. For example, we can make good use of the pace analysis to observe the angle variation of patients' feet before and after the injury, keep pace with situations of slower paces towards those with COPD disease, or record the exercise scenarios of some athletes. In this paper, we use video cameras to extract the human body outline and take advantage of the outline distance map measured between the gravity center and human body outline to establish the human body model including the head, upper body, lower body, hands and foot. What's more, we take the gravity center of these 7 different parts as our interest points. During the walking behaviour, we can obtain the angle and pace information of body parts to do the pace analysis by using these interest points and know if there is any abnormality for the health condition of subjects. In the end, we do the pace analysis with our proposed algorithm using cameras in the embedded system and receive good experimental results as shown on the C environment.