Research on Gait Evaluation Method Based on Machine Vision

Gait is a movement posture that appears during walking, which can reflect the health of the human body. Aiming at people with lower limb dyskinesia, a gait evaluation method based on machine vision is proposed. Through the posture estimation network combined with RGB-D data, complete the patient's three-dimensional joint point recognition. On this basis, the gait parameters are calculated through the motion trajectory of the joint points, and the Spearman algorithm is used to analyze the gait parameters in the normal population under different age and gender groups. As well as the performance between the normal population and the disabled population, select parameters that are not significantly related to age and gender, and show significant differences between the healthy population and the lower extremity dyskinesia population, and assign different weight coefficients to each parameter to establish information reflecting each parameter Comprehensive gait evaluation system.