Method, system and medium for three-dimensional measurement of foot shape based on multi-eye stereo vision based on color coding

Abstract

The invention discloses a method, system and medium for three-dimensional measurement of foot shape based on multi-eye stereo vision based on color coding. The method includes: using a multi-eye image collection platform to simultaneously photograph the foot shape wearing socks with color-coded patterns by using multiple cameras, Carry out image acquisition; obtain the internal and external parameters of the camera according to the collected images and the multi-camera calibration algorithm based on the sphere; perform intensive feature point matching of the foot shape according to the internal and external parameters of the camera; perform Model construction and foot shape parameter measurement. The invention has the advantages of low cost, easy identification and matching of feature points, high density of generated point clouds, fast calculation speed, etc., and accurately and completely obtains the foot model and related size parameters.

Description

technical field [0001] The invention relates to the technical field of foot shape measurement, in particular to a color coding-based multi-eye stereo vision three-dimensional foot shape measurement method, system and medium. Background technique [0002] At present, the non-contact technologies commonly used in 3D foot measurement include methods based on laser scanning reconstruction, methods based on structured light reconstruction and methods based on stereo vision reconstruction. Most of the foot shape three-dimensional measurement systems on the market are based on the method of laser scanning and reconstruction. Line lasers are used to form a closed plane to scan the feet. At the same time, the camera is used for imaging to obtain the surface information of the foot shape on the section of the light belt, and then through the motor along the vertical direction. Exercise to get the 3D model of the foot shape. The measurement method based on structured light reconstruct...

AI Technical Summary

Problems solved by technology

However, the reliability of the measured parameters is lower compared with the method of directly reconstructing the model

The reconstruction method based on scattered spots is sparse in feature points, so the reconstructed model is also sparse.

At the same time, the measurement methods based on stereo vision reconstruction all require accurate camera parameters, and when the checkerboard calibration method is applied to multiple cameras with a wide distribution, the camera parameters are easy to introduce large cumulative errors in the process of relationship transfer

Images