宋仙丽 夏建磊

DOI：10.19850/j.cnki.2096-4706.2021.08.027

摘  要：图像匹配是人体测量系统实现围度测量的关键。图像匹配的基础和前提是围度标记点的精准识别，标记点的识别效果受颜色、形状等因素的影响。在图像匹配过程中，可以通过提高彩色标记点的识别效率来提高图像匹配准确性。文章设计多种不同颜色和不同形状的标记点，利用基于HSV模型的颜色空间聚类算法对标记点进行各颜色的分类和坐标的聚类。经实验验证，设计的黄品青圆形的标记点类型的识别效果较好。

关键词：标记点识别;颜色空间聚类算法;HSV模型;图像匹配

中图分类号：TP391.4    文献标识码：A    文章编号：2096-4706（2021）08-0095-05

Optimization Design of Girth Marker Point Based on Color Space Clustering Algorithm

SONG Xianli，XIA Jianlei

（School of Electronic and Information，Zhongyuan University of Technology，Zhengzhou  450007，China）

Abstract：Image matching is the key for human body measurement system to realize girth measurement. The basis and premise of image matching is the accurate identification of girth marker point. The identification effect of marker point is affected by color，shape and other factors. In the process of image matching，the accuracy of image matching can be improved by improving the identification efficiency of color marker point. In this paper，a variety of marker points with different colors and shapes are designed，and the color space clustering algorithm based on HSV model is used to carry out color classification and coordinate clustering on the marker points. It is verified by experiment that the recognition effect of the designed yellow-magenta-cyan circle mark point type is better.

Keywords：marker point identification;color space clustering algorithm;HSV model;image matching

0  引  言

随着数字技术的出现，个性化服装定制已成为当代消费者新的追求^[1]，而服装设计也正朝着数字化、智能化的方向快速发展^[2]。准确的人体尺寸是服装定制的先决条件^[3]。人体测量技術从开始的人工接触式测量发展到现在的非接触式测量^[4]。非接触式测量又分为主动式和被动式两种。主动式设备往往价格昂贵、体型庞大且有辐射^[5]。被动式测量利用普通相机采集人体彩色图像，之后对人体图像进行一系列图像处理，获得被测者的尺寸信息。基于多视点立体图像的人体围度测量系统^[6]在测量围度时，其定制的紧身衣测量部位镶嵌着一些循环使用的彩色标记点，图像匹配是该系统实现围度测量的关键。图像匹配的基础和前提是标记点的精准识别^[7]，标记点的识别效果又受颜色、形状等因素的影响。……