一种融合的聚类算法在服装图像颜色识别上的应用  

Application of a Combined Clustering in Color Recognition of Clothing Image

作　　者：张轩宇 邓作杰[1] 高樱萍 黄威 ZHANG Xuan-yu;DENG Zuo-jie;GAO Ying-ping;HUANG Wei(College of Computer and Communication,Hunan Institute of Engineering,Xiangtan 411104,China)

机构地区：[1]湖南工程学院计算机与通信学院,湘潭411104

出　　处：《湖南工程学院学报（自然科学版）》2020年第4期51-54,共4页Journal of Hunan Institute of Engineering(Natural Science Edition)

基　　金：湖南省教育厅重点资助项目(18A343).

摘　　要：颜色是服装图像的重要特征,服装图像颜色识别广泛地应用在服装图像搜索中,现有的颜色识别方法都是把颜色特征粗略地划分为几个具体的值来进行匹配,会导致颜色识别出现比较大的误差.为了解决这种问题,提出了一种融合均值平移和k均值聚类算法的颜色识别方法,首先使用k均值聚类算法对服装图片的环境和布料颜色进行提取和聚类,然后对图像颜色进行重建,最后对重建后的图像颜色使用均值平移进行处理,它对像素颜色的聚类中心距离进行计算,并将其距离同含有标准颜色数据库进行比较来识别服装图像的颜色.实验结果表明,融合均值平移和k均值聚类的颜色识别方法可以有效地对服装图像颜色进行识别.Color is a key feature of clothing image.Color recognition of clothing image is widely used in clothing image search.The existing methods of color recognition roughly divide the colors into several values for matching,where the error is big in color recognition.To solve this problem,this paper presents a method of color recognition based on k-means and mean shift.By using this method,the background and foreground of clothing image are extracted and clustered;the color of clothing image is then reconstructed,and the reconstructed color of clothing image is processed by mean shift.The distance from the clustering center of the pixel colors is calculated and compared with the distance of the standard colors in database.The experiment results show that the method based on k-means and mean shift can effectively identify the color of clothing image.

关 键 词：均值平移 K均值聚类 颜色识别 图像特征 

分 类 号：TP312[自动化与计算机技术—计算机软件与理论]