Treadmill and vision system for human gait acquisition and analysis

Abstract

This paper presents a developed low cost system for human gait analysis. Two web cameras placed in opposite sides of a treadmill are used to acquire images of a person walking at different speeds on a treadmill, carrying a set of passive marks located at strategic places of its body. The treadmill also has passive marks with the color chosen to contrast with the ambient dominant color. The body joint angle trajectories and 3D crossed angles are obtained by image processing of the two opposite side videos. The maximum absolute error for the different joint angles acquired by the system was found to be between 0.4 to 3.5 degrees. With this low cost measurement system the analysis and reconstruction of the human gait can be done with relatively good accuracy, becoming a good alternative to more expensive systems to be used in human gait characterization.