Subdivision And Discrimination Of Upper Body Features For Young Female

Under the background of intelligent development of clothing industry and personalized clothing consumption,personalized human body model has become a hotspot in the field of multidisciplinary interdisciplinary application research.In order to obtain the key parameter information of personalized human body model,the aim of this paper is to obtain personalized human body structural parameters based on the body surface feature points,through the subdivision of the tripartite body surface characteristics(width,thickness,and height)to describe the morphological differences of young females'upper body and establish a discriminant model.Human body shape feature points are the key information and basic units to construct the structure of human body model.The work took 141 female college students in central China as the object and collected a total of 28 2D morphological feature data(breadths,thicknesses,and heights)of upper body related to 12 feature points by photo measurement method.According to the results of correlation analysis between the morphological characteristic variables,the feature variables of breadths,thicknesses,and heights significantly correlated with the variables of their coordinate orientation respectively.Then,the 5 feature factors of upper body shapes were extracted and interpreted as the factors of shoulder-neck thickness,height,bust-waist thickness,shoulder-neck breadth,and bust breadth with characteristic value over 1 by the principal component analysis,and a total of 9 variables were selected from these morphological feature factors and summarized into 6 upper body typical indices in three dimensions of young female,which subdivided into 3 categories by K-means clustering method respectively,and the subdivided indices merged into 3~6 compound types characterizing the differences of upper body shapes.Finally,the typical feature recognition equations established by the Bayes discriminant analysis.Furthermore,according to the relationship between the typical indicators and other variables determined the second-order and third-order indices of morphological features,and corresponding regression(recognition)equations for feature prediction are established.Fifteen new samples were used to verify the validity of the upper body morphological feature discrimination model.By subdividing typical morphological characteristics indices,a morphological feature database was established to distinguish the physical differences of young females,and the differentiated feature parameters were provided for individual feature recognition.On the other hand,the classification system of morphological feature discrimination is used to obtain three-dimensional coordinate information of body surface feature points and construct the structure model of the upper body of an individual based on it.The construction of the body surface of an individual provides key and comprehensive characteristic parameters and is suitable for the field of personalized customization,that is more suitable to applied in the domain of personalized customization.