Font Size: a A A

Verification Of Waist Section Graphics Based On Non-contact Anthropometric Measurement

Posted on:2020-02-10Degree:MasterType:Thesis
Country:ChinaCandidate:Y Q NingFull Text:PDF
GTID:2381330578979229Subject:Textile Science and Engineering
Abstract/Summary:PDF Full Text Request
With the development of science-technology and big data,intelligentization has become the general direction in the development of the times.The combination of garment industry and intelligent technology has become an inevitable trend.The non-contact measurement technology which is efficient caters to today's market.It should be used widely to promote the development of the garment industry.However,many companies aren't willing to use non-contact measuring equipment,because the equipments are too expensive,people also doubt about the data and graph which generated from it.Therefore,studying on graphic data obtained by automatic non-contact measurement and verifying its accuracy is an urgent issue.Solving the issue will provide correspondingly basis and help for improving the accuracy of intelligent equipments.It will also accelerate the intelligent development process of the garment industry.As the important control part size of the human body,the circumference dimension plays an pivotal role in the design of the garment structure.Therefore,this paper selects the waist curve graph obtained by the non-contact measurement technology to study.Under the standard and basis of the manual measurement data,the standard waist curve model is established.By verifying the waist data and graph obtained by non-contact measurement,a method that verify whether the graphics obtained by the non-contact measurement is available or not was found.The specific research contents are as follows:(1)Experiment on the collection of human body size:In this experiment,208 female students aged from 18 to 25 years old and female employees aged from 30 to 45 years old were selected as experimental subjects.Traditional manual measurement and non-contact anthropometric measurement were used to collect the size of the relevant research parts of human body.Then,dealing the measured data with basic statistical analysis.(2)Body cluster analysis:Under 7 body classification indices(body to chest ratio,chest and waist difference,chest to waist ratio,hip and waist difference,hip to waist ratio,hip and chest difference,hip to chest ratio),all samples were analyzed by body cluster analysis.Then,multivariate correlation analysis was performed on the 7 body type indices and the size of the four human body parts(height,chest circumference,waist circumference and hip circumference).The chest to body ratio was selected as the basis of the body cluster analysis according to the analysis results.Finally,women of 155/80A and women of 160/92A were selected for the following study.(3)Establish a figure standard waist circumference curve graphic and a model for non-contact measurement of waist circumference curve:firstly,the characteristic value data of the waist of subjects were collected with Martin ruler,vernier caliper,marker belt and other experimental tools.Then,the collected data were programmed with MatLab software to obtain the fitting point coordinates of each point on the waist curve.Finally,the standard waist circumference curve was generated in AutoCAD software,and the mathematical model of fitting waist circumference curve with double elliptic curve was obtained by using the circumference calculation formula of ellipse.Imageware software was used to extract the waist circumference cross-section figure from the body point cloud model automatically generated by the Size Stream non-contact body scanner and save it as.dxf format file.Using AutoCAD software to obtain waist circumference curve and then correcting and smoothing it.The characteristic values of waist width and waist thickness were extracted after correcting the waist circumference curve,and a mathematical model of the non-contact waist curve was fitted with a double elliptic curve by using the circumference formula of ellipse.(4)Data validation analysis:The standard waist circumference curve was fitted by the method of double elliptic curve fitting,and then the waist circumference was obtained after fitting.Based on the manually measured waist circumference,T-test analysis was performed on the fitted waistline.If the analysis result indicates that there is no significant difference between the fitted waist circumference and the manually measured waist circumference,the fitted waist circumference is available.Therefore,the fitting method of the double elliptic curve will continue to be used to fit the waist circumference of the non-contact measurement,and then the fitting value of it and the standard waist circumference curve will be verified and analyzed to determine whether the data obtained by non-contact measurement is available or not.(5)Verification of non-contact measurement waist circumference curve graph and verification method:Matlab was used to calculate the superposition area S of standard waist circumference curve graph and non-contact measurement waist circumference curve graph,and the critical value range of superimposed area S was determined by the shift of waist and the shift of waist thickness.If S is less than the critical value,it means the waist curve obtained by the non-contact measurement is roughly consistent with the trend of the standard waist curve and they have little difference,and the figure is available.In order to test the feasibility of the graphic verification method,the two types of experimental samples both selected 5 samples to be as the verification samples.According to the analysis results of each sample,the conclusion was made to judge whether the graphics were available or not,so as to verify the feasibility of the graphic verification method.
Keywords/Search Tags:non-contact anthropometry, waist circumference curve, curve fitting, model construction, graphic verification
PDF Full Text Request
Related items