Method and system for analyzing joint stress during single-leg sprint in fencing

Abstract

The application discloses a fencer single-foot sprint joint stress analysis method and system, and relates to the technical field of machine learning, comprising: acquiring key video frames of a to-be-analyzed athlete during a fencer single-foot sprint; performing feature extraction on the key video frames based on a trained regional convolutional neural network to obtain limb segmentation data of the to-be-analyzed athlete during the fencer single-foot sprint, wherein the limb segmentation data contains joint features; constructing a target joint model of the to-be-analyzed athlete based on the limb segmentation data and adult human body inertia parameters; performing stress analysis on joints of the to-be-analyzed athlete based on the target joint model to obtain joint stress and torque equations; and calculating joint stress and joint torque of the to-be-analyzed athlete during the fencer single-foot sprint based on the joint stress and torque equations. The application alleviates the technical problems of high cost, prior preparation and hysteresis in the prior art.

Images