Measurement of knee flexion/extension using a 2-D markerless technique

Several measurement tools are typically used in assessing knee flexion/extension: electro-goniometers, inertial sensors, 2-D and 3-D marker based motion analysis systems [1]. Most of the traditional clinical analysis methods require applying sensors to the patients, which may cause discomfort and hinder the natural movement. To overcome the aforementioned limitations, in this study, we propose a 2-D markerless technique to measure knee flexion/extension. This study is an enhancement of our previous markerless technique [2]. The focus of the first study was the analysis of the 2-D kinematics of the shank and foot complex using a single lateral view by defining a three-segment model and tracking its movement using selected patches rigid to the segment. To simplify the problem, a black background was used and the subjects wore long black socks to cover the contra-lateral leg. In this study, we developed a markerless technique, which does require neither a black background nor a black long sock on the contra-lateral leg. The presence of a more complex background was dealt with by using an adaptive statistical background subtraction model, while the removal of the long black sock on the contra-lateral leg resulted in occlusions that are dealt with by defining an additional patch per segment to be used when the occlusion occurs on the main patch. As for the previous study, the proposed technique will be validated acquiring the same trials with both the single camera used in the proposed markerless technique and a traditional stereophotogrammetric system. This technique is expected to provide a simple and reliable knee flexion/extension measurement.