Height Measurement Method Based on Skeleton Breakline Ridge Regression

Abstract

The invention discloses a height measurement method based on skeleton broken line Ridge regression. By calibrating the human skeleton in the scene, the measurement of human height is realized. Accordingly, the human skeleton model is used to select and extract the key points of the human height skeleton in the captured image. The obtained coordinates of the key points of the criminal's height skeleton are used to calculate the distance shape parameter and direction shape parameter of the height skeleton by using the 2-norm and the arctangent function. Then, based on the pose nearest neighbor matching of the included angle of the skeleton, the distance shape parameters of the reference human height skeleton are extracted. Finally, a height measurement prediction model is constructed based on the training data set of the skeleton broken line height model and the analytical solution of the Ridge regression loss function to realize the height measurement of criminals. The invention has the advantages of simple and easy operation, strong reliability, high scene adaptability, etc., and can also complete the height measurement when the human body in the image is distorted or has a complicated posture.

Description

technical field [0001] The invention belongs to the technical field of height measurement, and more specifically relates to a height measurement method based on skeleton broken line Ridge regression. Background technique [0002] Visual measurement technology is based on machine vision, integrating optoelectronics, computer technology, laser technology, image processing technology and other modern science and technology into one, forming a comprehensive measurement system of light, machinery, electricity and calculation, with non-contact, full field of view Features of measurement, high precision and high degree of automation. According to whether the camera needs to be calibrated in advance, visual measurement can be divided into measurement under calibration and measurement without calibration. [0003] The measurement under calibration is a common method of visual measurement, which requires the use of high-precision special calibration blocks to determine the internal p...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to propose a height measurement method based on skeleton broken line Ridge regression, which overcomes the interference of background information and complex structure mixed information in the captured image of height measurement, and solves the situation that the human body in the image is distorted or has complex postures. Under the problem of not being able to measure height

Images