基于图像的非接触式人体参数测量方法  

Image-based non-contacting anthropometric method

作　　者：邹昆[1] 马黎[2] 李蓉[1] 徐晨 ZOU Kun;MA Li;LI Rong;XU Chen(School of Computer Engineering, University of Electronic Science and Technology of China, Zhongshan Institute,Zhongshan 528402, China;School of Computer Science and Engineering, University of Electronic Science and Technology of China, Chengdu 611731, China;HD Video R &?D Platform for Intelligent Analysis and Processing in Guangdong Engineering Technology Research Centre of Colleges and Universities (No. GCZX-A1409), Shenzhen 518060,China)

机构地区：[1]电子科技大学中山学院计算机学院,广东中山528402 [2]电子科技大学计算机科学与工程学院,四川成都611731 [3]广东省高校高清视频智能分析与处理工程技术研究中心(No.GCZX-A1409),广东深圳518060

出　　处：《计算机工程与设计》2017年第2期511-516,共6页Computer Engineering and Design

基　　金：国家自然科学基金项目(61502088,61300095);广东省高等学校优秀青年教师培养计划基金项目(Yq2013206,Yq2015241,Yq2015242).

摘　　要：在对现有的测量方法进行研究总结后,一种人体参数测量方法被提出。通过人脸检测从人体正、侧面图像中定位人脸特征区域,结合人体比例和其它先验知识对人体特征区域、特征点进行定位,由特征点之间的距离拟合得到人体参数。实验结果表明,该方法能够快速准确地通过图像得到人体参数。After reviewing the existing measuring methods,an image-based non-contacting anthropometric method was pro-posed.Face detection was used to locate the face regions in the front and side images of human body.The positions of feature re-gions and feature points were determined by utilizing the proportions of human body and some other prior knowledge.The nee-ded human body dimensions were calculated based on the distances between feature points.Experimental results show that the body dimensions can be quickly and accurately acquired from the images using the proposed method.

关 键 词：非接触式测量 人脸检测 先验知识 特征区域 特征点 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]