摘要
以190名青年男大学生为研究对象,通过照片测量法、三维人体扫描法获取人体正侧面照片及三维点云数据。基于三维点云数据,提取颈肩部轮廓曲线进行椭圆傅立叶拟合得到椭圆系数,经主成分分析和聚类分析,将青年男性颈肩部正侧面形态各分为3类。采用角度半径和欧几里得距离实现二维照片与三维点云数据之间的匹配,并验证匹配准确率。结果显示:颈肩部正面轮廓匹配验证准确率达92.5%,侧面轮廓匹配验证准确率达95.0%。该研究为基于人体照片的人体形态研究提供了理论依据。
This study took 190 young male college students as the research objects,obtained the front and side photos and 3D point cloud data by photo measurement and 3D scanning.Based on the 3D point cloud data,the contour curves of the neck and shoulders were extracted for elliptic Fourier fitting to obtain the ellipse coefficients.After principal component analysis and cluster analysis,the front and side shapes of the neck-shoulder of young men were divided into three categories.The angular radius and Euclidean distance were used to achieve the matching between the 2D photos and 3D point cloud data,and the matching accuracy was verified.The results show that the neck and shoulder frontal contour matching verification results accuracy rate is 92.5%,and the neck and shoulder lateral contour matching verification results accuracy rate is 95.0%.This study provides a theoretical basis for morphological research based on human photographs.
作者
张健
徐凯忆
赵崧灵
顾冰菲
Zhang Jian;Xu Kaiyi;Zhao Songling;Gu Bingfei(School of Fashion Design&Engineering,Zhejiang Sci-tech University,Hangzhou,Zhejiang 310018,China;Clothing Engineering Research Center of Zhejiang Province,Hangzhou,Zhejiang 310018,China;Key Laboratory of Silk Culture Heritage and Products Design Digital Technology,Ministry of Culture and Tourism,Hangzhou,Zhejiang 310018,China)
出处
《针织工业》
北大核心
2024年第9期71-76,共6页
Knitting Industries
基金
国家自然科学基金项目(61702461)
中国纺织工业联合会应用基础研究项目(J202007)
浙江省哲学社会科学规划艺术学课题资助项目(24NDJC171YB)
浙江理工大学科研业务费专项资金资助项目(24076114Y)
浙江理工大学优秀研究生学位论文培育基金项目(LW-YP2021054)。