Abstract: The aim of this study is to develop a new method to calculate the characteristic section of the waist and hip without clothing pressure by using 3D scanning and finite element simulation technology. This goal contributes to a deeper understanding of morphological characteristics in the human body’s natural state and its interactions with tight-fitting clothing, advancement of the development of personalized clothing design and manufacturing, and enhancement of the wearing comfort and overall well-being.
To achieve this goal, based on the central limit theorem in statistics, a sample of 30 young women aged between 20 and 25 years old was selected for 3D scanning. Then, through reverse engineering technology, the characteristic sections of the waist and hip under clothing pressure and without clothing pressure were extracted. On this basis, a pressure iteration algorithm based on finite element method was developed, which can simulate the clothing pressure effect of tights on the waist and hip sections, and then calculate the non-uniform clothing pressures and body surface displacement vector caused by tights. In addition, the high-and low-level output response regression method was used to comprehensively consider three groups of main factors including human parameters, tight size and section shape characteristics, and to establish a regression model to predict the pressure distribution and displacement. Finally, goodness of fit (R²) and the global comparison hashing algorithm were used to evaluate the validity of the model.
The pressure iteration algorithm based on finite element simulation can calculate the pressure and displacement vectors for 120 key points in the characteristic sections of waist and hip. Through the five-level and two-factor experimental design, it is found that the factors such as waist and hip circumference, body fat percentage, waist and hip circumference of tights, curvature and soft tissue thickness have important effects on the pressure and displacement of tights. Accordingly, a regression model was constructed to predict the clothing pressure and body surface displacement of tights. The empirical test shows that the pressure goodness of fit of the model reaches 89.73% on average, and the displacement vector goodness of fit is also as high as 85.21%. Furthermore, the characteristic sections of the waist and hip without clothing pressure were compared with the section without garment compression obtained by scanning. The average shape similarity is 85.28%, and the standard deviation is 4.99%, which confirms the accuracy of the method in restoring the human shape without garment compression.

Key words: waist and hips, characteristic section, pressure iteration algorithm, finite element simulation, clothing pressure, displacement vector

摘要： 为实现从着紧身裤腰臀部特征截面反算未着装腰臀部特征截面，采用三维扫描仪获取30名女体腰臀部数据，提取腰臀部特征截面，搭建基于有限元方法的压力迭代算法，拟合非均匀服装压和位移矢量，从而构建腰臀部特征截面的输出响应回归模型。研究表明，压力模型拟合优度均值为89.73%，位移矢量拟合优度均值达到85.21%，反算得到的未着装特征截面，与实际未着装特征截面的形状相似度平均为85.28%，标准差为4.99%。研究可为三维测量得到真实人体形态提供参考。

关键词: 腰臀部, 特征截面, 压力迭代算法, 有限元仿真, 服装压, 位移矢量