| With the improvement of living standard,people's requirements for wearing clothes are also gradually improved,and personalization and comfort are paid attention to,so,the garment customization industry will usher in new changes.However,now the customization of clothing tends to be high-end,and there are specialized quantity and body division to measure it,so,the cost is very high.Although optical scanner is also used in the measurement of human body size,but it is not widely used,according to the above problems,this paper designed a human body parameters testing system used in the clothing customization industry,and verify the reliability of the system in human feature size extraction.The specific research work is as follows:(1)A human dimension measuring instrument is designed according to the requirements of human dimension acquisition.The measuring instrument mainly includes non-contact measuring instrument which is used to obtain data through 3D scanner and contact measuring instrument which is used to read body size through e ruler.This method can be used to measure human with clothing and solve the problem that the optical human body measurement needs to be replaced by the corset.At the same time,the software system of human dimension extraction is developed,design the program through C++ and OpenGL,and develop the software interface through Qt.(2)Geomagic Studio,a professional reverse-engineering software,is used to conduct noise reduction,hole filling,uniform sampling and other operations on the obtained point cloud model,and use Geomagic Qualify to fitted model for 3D deviation analysisthen,the triangular facet model is generated and output as OBJ standard 3D format file.To realize the unification of local coordinate system and world coordinate system and calculate the formula of coordinate transformation to make the movement of human model.(3)An efficient slicing algorithm is designed for the model,and the point cloud coordinates and the number of connected domains are extracted by using the algorithm;An algorithm for the extraction of human body feature points was proposed.The number of connected domains obtained was used to determine the position of human body bifurcation points,and the location of coordinate extreme feature points was determined through the extremum of dimension size.At the same time,the preliminary segmentation of human model is realized according to the demand.Then,convex hull method is used to calculate the size of the circumference and the distance between the feature points to calculate the size of the length.At the same time,the preliminary segmentation of the mannequin was realized according to the requirements.(4)The repeatability,stability and practicability of the human parameter test system are verified through the parameter extraction experiment.The stability of the human parameter test system meets the maximum allowable floating value(no more than 1.5cm)requirements of the ANSUR project in the United States;The maximum measurement error of the tightness garment is 5.40 mm,and the maximum measurement error of the wearing shirt and pants is 10.09mm,which basically meets the specified error of GB/T 23698-2009 General Requirements for 3d Scanning Anthropometric Measurement Methods and the error given by Dekker.L(0.7-1.7cm).Finally,the error is analyzed,and Most of these errors can be optimized by adjusting the software algorithm and increasing the precision of the model. |
PDF Full Text Request