摘要
针对三维人脸识别中单一特征信息不足,采用一种基于整体信息和局部信息相融合的识别算法,以提高识别率。首先将预处理的三维点云用多层次B样条曲面拟合,获取精确的人脸曲面拟合函数,将控制点映射为深度图像,并根据人脸曲面函数和生理特征提取过鼻尖的中分轮廓线和水平轮廓线;其次对深度图像采用二维主元分析(2D-PCA)算法提取整体信息,对轮廓线采用改进的ICP算法匹配,作为局部信息;最后用加权求和法在决策级进行信息融合。采用CASIA3D人脸库完成识别测试,实验结果表明,本文算法明显优于单一特征信息下识别算法,且对姿态有较好的鲁棒性,同时不增加算法复杂度。
In terms of the scarcity of single feature information in 3D face recognition,a recognition algorithm based on the fusion of global and local information is proposed in this paper for improving the recognition accuracy.Firstly,we make a surface fitting on the preprocessed 3D cloudy points by using multilevel B-spline and acquire an accurate face surface fitting function.And then the control points of the function are mapped into the range image,and the central profile and the horizontal outline cross the nose top are extracted according to the surface function and physiological characteristics of the face.Secondly,2D principal component analysis（PCA） is applied on the range image for extracting global information,and the contours are matched using modified ICP algorithm as local information.Finally,the weighted sum method is used to achieve information fusion at decision stage.The recognition test is executed on CASIA 3D face database,and experimental results demonstrate that the method is obviously superior to the methods with single feature information and it is also robust to posture without increasing algorithm complexity.
出处
《光电子．激光》
EI
CAS
CSCD
北大核心
2012年第10期1977-1982,共6页
Journal of Optoelectronics·Laser
基金
中央高校基本科研业务经费资助(CDJXS12160004)资助项目
关键词
人脸识别
整体信息
局部信息
信息融合
face recognition
global information
local information
information fusion
