A method for automatically detecting the height of a human body

Abstract

The application discloses a method for automatically detecting human height, and relates to the technical field of intelligent measurement, comprising the steps of projecting coded structured light to a target human body area based on a structured light projection module; the application obtains a three-dimensional point cloud of a human body surface through structured light projection, combines a kinematic state space model and limit stretch state recognition, and inversely deduces an upright height based on a skeleton segment length invariable constraint reverse deduction theory; a human body kinematic state space model is introduced, joint angles, angular velocities and skeleton segment lengths are taken as state variables, physiological constraints are introduced for joint optimization, a kinematic state space model and multi-constraint optimization are combined, a head height vertex value is taken as a limit stretch state based on gait cycle analysis and recognition, the extracted skeleton segment length is taken as an invariable constraint, limit stretch state recognition and theoretical upright posture conversion are combined, a time-varying state space model and motion constraints are combined to predict the state during the shielding period, and continuous skeleton state acquisition during the shielding period is realized.

Images