Abstract:

Clothing patterns is an important feature of the She nationality clothing, so it has important significance to apply digital research based on the characteristics to the computer automatic classification of She nationality clothing. In this paper, 12 typical She nationality clothing were selected , feature key points were abstracted from the clothing by using the Gauss fuzzy and SIFT method, and the key points distribution of different parts such as neckline, placket, cuffs and side slits were acquired. The results show that, there is good agreement between the key point distribution obtained by computer and the different categories of the clothing feature. The key design area and sorting of different She nationality clothing were acquired by the analysis on the key points quantities. The complexity of the clothing texture can be characterized by the vector variance of key points. This method can effectively identify the different categories of She nationality clothing and provide a reference for the clothing design.

Key words: She natinality clothing, Gauss fuzzy, scale-invariant feature transform, clothing feature, key points distribution, clothing category