基于京剧脸谱意象色彩的服饰纹样自动配色

为提升服饰纹样色彩配色效能,利用自适应颜色聚类及自动配色机制,实现基于意象场景的服饰纹样自动配色。以京剧脸谱为范例,搜集了三类角色共计150幅脸谱图像作为源图,采用分割、降噪等操作获取脸谱主体内容;利用二分K-均值聚类算法,对单一样本进行逐一自适应色彩提取,在此基础上进行二次聚类获得各角色脸谱提取色色值、占比及共现比率等特征参数;最后设计融合颜色聚类数、共现比率及目标区域特征要素的动态配色机制,开发色彩解析和纹样自动配色软件。实验结果表明,配色控制参数、源图类型、纹样形态等因素影响选色顺序及最终配色效果,实现基于意象场景的纹样色彩自动迁移,为当代服饰纹样色彩设计提供决策参考。

Automatic coloration of clothing pattern based on color parsing of Peking Opera masks

In order to improve the coloration efficiency of clothing pattern design, an automatic color matching mechanism based on adaptive color clustering of scene images was proposed. 150 images of Peking Opera masks from three roles were collected as the source samples. The Peking Opera mask body of each sample image was obtained by image segmentation and denoising. The bipartite K-means clustering algorithm was utilized to extract adaptively the extracted colors of each sample. In addition, the secondary clustering was carried out to acquire the main color value, proportion and co-occurrence ratio of Peking Opera mask images. A dynamic color matching mechanism integrating the number of color clusters, co-occurrence ration and structure characteristics of the target pattern was designed. Consequently, an automatic color parsing and matching software package was developed. Results confirm that the color matching control parameters, source images, pattern shape and other factors affect the color selection sequence and the final matching effect. The proposed method can quickly achieve the pattern color migration based on the source images, and can provide reference for auxiliary decision-making on clothing pattern color design.