Fuzzy Logic Evaluation of Knee Flexion Angle During Gait

In this paper, we propose a method for quantitative gait analysis with videos. First, the joint position coordinates are estimated with the gait video. Next, by tracking the person and analyzing the knee joint, we obtain the time series data of the subject's knee flexion angle. Finally, based on fuzzy logic, we evaluate how close the subject's gait was to a normal gait. As a result, the mean value of fuzzy degree is 0.70±0.031(range: 0.64-0.74) for healthy adults and 0.18±0.043(range: 0.11-0.22) for patients. This system enables quantitative evaluation of gait more easily than existing methods.