Video based Gait analysing system

Introduction: Walking is the first way of displacement for humans and essential for daily life activities and social participation. The human gait can be analyzed from several points of view and specialties. Most gait capture systems use direct measurement techniques to acquire specific motion information, but the subject’s natural movement get hindered due to the presence of cables and other components. Thus, this proposal suggests a video-based system where the coordinates of the markers were obtained using the camera, passive markers, and Matlab. Methods: In this system, input is acquired by a digital camera. Five white passive markers are placed on different joints of the lower limb and then the movement of these markers are captured as video input and is processed to detect the position variations. This position variation is represented in the form of a stick diagram derived by connecting all the joint points. Then by using a cosine formula the angle variation at each of these joints are computed. Results & Discussions: The angle variations of the complete gait cycle of each joint are obtained for each frame and are graphically plotted. This helps to directly visualize the angle variation at the time of processing itself. This algorithm is applied in different gait patterns for male and female and the people with different heights and is compared with the standard values available in the literature. During the analysis, each joint angle measurements has plotted separately. Conclusions: From the analyzed kinematic gait parameters, the measures of hip angle, knee angle, and toe angle are obtained. The main advantage of this algorithm is the reduction in complexity and time consumption. This system helps to quantify the gait parameter that helps the healthcare professional to analyse the abnormalities in gait pattern and its treatment planning. Key words: Gait analysis, Walking, Video based, Hip angle, Knee angle References 1. Diego Andres Guffanti, Alberto Brunete, and Miguel Hernando. Non-invasive multi-camera gait analysis system and its application to gender classification. IEEE Access, PP:1–1, 05 2020. 2. Chandra Prakash, Kanika Gupta, Anshul Mittal, Rajesh Kumar, and Vijay Laxmi. Passive marker-based optical system for gait kinematics for lower extremity. Procedia Computer Science, 45:176185, 2015. 3. Rolf Moe-Nilssen. A method for reliability analysis of speed-related repeated measures gait data. Gait & Posture, 33(2):297299, 2011