Abstract: How to achieve precise body shape analysis is a key issue to meet consumer requirement on apparel fit. In order to improve the fitting of garment on hips, this paper focused on the differences of curves about hip sections to study on subdividing of young female hip shapes by statistical analysis. Measurement of 300 young women aged from 20-24 were taken with 3D-body scanner; the position data relating to the hips was conducted using cluster analysis, then extracted typical index of each type by comparing their values, and picked up 6 characteristic indices contain almost information of hip shape at the same time; according to characteristic indices, the female hip shapes were subdivided into 5 types through analysis of variance; the rationality of subdividing into 5 types was tested and verified by the mean of comparing with curves of each type hip sections at last.

Key words: 3-D body scanning, hip, typical index, cluster analysis, subdivide of body shape