服装品牌与服装廓形分类
Clothing Brands and Clothing Silhouette Classification

背景与目的：廓形是服装设计的重要元素。设计师通过廓形线条中的设计变化来表达灵感。本研究的目的是构建一个廓形分类标准，并对服装品牌进行廓形分类。研究方法：实验服装图片选自国际一线品牌：传统的法国奢侈品牌香奈儿、日本潮流服装品牌川久保玲和意大利奢侈品品牌瓦伦蒂诺。实验数据是基于服装制图主要部位，选取其中8个的测量结果用于实验分析。创新点：该分析结合了系统聚类和K-means聚类分析。实验结果得到了七种廓形类别，并且通过对廓形进行数字化分析和分类处理，来直观地展现品牌特征廓形。结论：该方法可用于未知服装廓形的自动分类，也可为未来服装设计提供参考。
Outline is an important element in clothing design. Designers express inspiration through design changes in silhouette lines. The purpose of this study is to construct a silhouette classification standard and classify clothing brands’ silhouettes. The pictures of experimental clothing are selected from international first-line brands: traditional French luxury brand Chanel, Japanese fashion brand Comme des garcons and Italian luxury brand Valentino. The experimental data is based on the main parts of clothing drawing, and 8 measurement results are selected for experimental analysis. This analysis combines system clustering and K-means clustering analysis. The experimental results obtained seven types of silhouette categories, and through digital analysis and classification processing of silhouettes, the brand’s characteristic silhouettes were visually displayed. This method can be used for automatic classification of unknown clothing contours and can also provide reference for future clothing design.