Pedestrian foot recognition method applied to walking aid rehabilitation robot

The invention discloses a pedestrian foot recognition method applied to a walking-aid rehabilitation robot, and aims to solve the problem that when a low-cost RGB-D camera of the walking-aid rehabilitation robot only collects the foot postures of the lower part of the human body, foot recognition and pose analysis are carried out only by using the features of the feet of a pedestrian. A foot recognition method based on ellipse fitting is designed, constraints of ellipse long and short axis ratio, area and color similarity are increased, and a left and right foot classification method is designed. And compared with calibrated poses, errors are calculated, and the engineering requirements are effectively judged and identified. The pedestrian foot recognition method applied to the walking-aid rehabilitation robot can be applied to pedestrian foot image recognition of the walking-aid rehabilitation robot, and the real-time performance and accuracy of image recognition are improved. 本发明公开了一种应用于助行康复机器人的行人脚部识别方法，针对助