Measurement and classification of shanghai female face shape based on 3D image feature

Define faces temporal width, bizygomatic breadth, mandibular breadth, maxilla-chin breadth, physiognomic facial length as the key characteristic parameters of face classification, which based on anthropometry, image feature and mathematical statistics theory. Characteristic data of face shapes for 201 cases of shanghai female college students are extracted through 3D non-contact image measurement. Young female in shanghai are divided into 8 kinds of face shapes by cluster analysis using SPSS software. And the proportion of various face shape is: round face and diamond shape face account for the largest proportion, heart-shape face account for the smallest proportion, melon seeds face, pear face and long face is the second, square face and ellipse face account for intermediate proportion. Finally, analysis of variance of each index and chi-square test of data prove reasonable selection of sample and facial index, scientific classification of human faces. The conclusions provide the engineering and aesthetics basis for anthropology and biomedical.