基于网络关系模型的畲族服饰色彩构成分析  

The color composition analysis of She costumes based on the nexus network model

作　　者：贾静 曹竟文 徐平华 林瑞冰 孙晓婉 JIA Jing;CAO Jingwen;XU Pinghua;LIN Ruibing;SUN Xiaowan(School of Fashion Design&Engineering,Zhejiang Sci-Tech University,Hangzhou 310018,China;Zhejiang Provincial Research Center of Fashion Engineering Technology,Zhejiang Sci-Tech University,Hangzhou 310018,China;MOC Key Laboratory of Silk Culture Heritage and Product Design Digital Technology,Zhejiang Sci-Tech University,Hangzhou 310018,China)

机构地区：[1]浙江理工大学服装学院,杭州310018 [2]浙江理工大学浙江省服装工程技术研究中心,杭州310018 [3]浙江理工大学丝绸文化传承与产品设计数字化技术文化和旅游部重点实验室,杭州310018

出　　处：《丝绸》2022年第11期100-107,共8页Journal of Silk

基　　金：国家自然科学基金青年基金项目(61702460);国家社会科学基金重点项目(19AMZ009);浙江理工大学科研业务费专项资金资助项目(2021Q057);服装设计国家级虚拟仿真实验教学中心项目(zx20212004);浙江省服装工程技术研究中心开放基金项目(2021FZKF05);浙江省教育厅一般科研项目(Y202250618);浙江理工大学教育教学改革研究重点项目(jgzd202202);浙江理工大学优秀研究生学位论文培育基金项目(LW-YP2021053)。

摘　　要：为区分不同地区畲族服饰色彩构成差异,文章利用图像分析技术,构建畲族服饰色彩网络关系,比对不同地区畲族服饰色彩构成情况。实验选取福建、浙江、江西三地畲族服饰共计300幅图像,利用改进的二分K-均值自适应聚类法对样本图像逐一聚类;在此基础上,对综合色图进行二次聚类,提取颜色占比、二元色共现率、色彩空间距离、色彩分散度等指标,构建服饰色彩网络关系模型。实验结果表明,三地配饰用色的色相视觉上较为相似,但用色、占比、色序及配色形式差异较大;单幅样本色彩提取耗时小于0.2 s,色彩网络关系模型输出平均耗时为1.4 s。该方法能够有效分析不同地区畲族服饰色彩特征及配色关系,以可视化的形式展示服饰色彩构成形态。Ethnic costumes are colorful and reflect profound cultural heritage.Parsing the color composition and coloration mechanism of ethnic costumes will help to enrich and improve the color system,inherit and regenerate the ethnic costume culture.The formation of national costume color is frequently linked to the geographical surroundings,living habits,ethnic culture and religious belief.Image analysis technology has been utilized to quantify the color characteristics of national costumes and parse the color rules,which can broaden the scope of color regeneration services in contemporary art,design and cultural communication.She nationality is a Southeast Chinese ethnic group with distinctive clothes and accessories.With the nexus network model as the analysis tool,this paper utilizes image analysis technology and selects a total of 300 clothing images of the She nationality in Fujian,Zhejiang,and Jiangxi as samples to quantitatively analyze their colors and describe the differences in different regions.Firstly,the selected image samples are preprocessed.Secondly,in order to avoid over-balanced color of fixed clustering centers,ensure the accuracy of basic color extraction and obtain relatively accurate color center value and cluster size,an improved bipartite K-means adaptive clustering algorithm is used to cluster the sample images for the first time,and the integrated color map is formed according to the color proportion in the experiment from the perspective of speed and convenience.On this basis,in order to facilitate the comparison of the color quantity and coloration relationship in different regions,K-means clustering method is utilized to uniformly set the cluster size for secondary clustering of the comprehensive color map based on literature research and experience cognition.To analyze the differences in color composition of clothing of She nationality in Fujian,Zhejiang,and Jiangxi,the clothing color is decomposed into four color characteristic indexes,namely,color proportion,paired color co-occurrence rate

关 键 词：畲族服饰 自适应聚类 色彩 网络关系 模型 可视化 

分 类 号：TS941.26[轻工技术与工程—服装设计与工程]