温宏愿,刘小军,刘增元

• 1:南京理工大学泰州科技学院

摘要(Abstract)：

传统识别方法主要通过谱回归和矩阵完整性约束,从带噪声的原始数据中得到干净的输入数据,达到提高识别人脸生物特征的目的,但忽略了低秩投影矩阵对识别带来的干扰,导致识别精度低、效率差的问题,故提出基于机器视觉的人脸生物特征改进识别技术。通过摄像机代替机器视觉对人脸图像进行采集,构建人脸图像双色反射模型,对人脸图像进行预处理,并以机器视觉为基础,结合局部差分二值模型(LDBP),实现对人脸生物特征识别的改进。实验结果表明,相比于传统识别技术,采用改进识别技术在进行人脸生物特征识别方面的识别精度较高,实用性较强,具有一定的优势。

关键词(KeyWords)： 机器视觉;人脸;生物特征识别;摄像机;双色反射;局部差分二值模型

Abstract:

Keywords:

基金项目(Foundation): 江苏省“333工程”科研资助项目(BRA2016177)~~

作者(Author): 温宏愿,刘小军,刘增元

DOI: 10.16652/j.issn.1004-373x.2018.10.041

参考文献(References)：

• [1]肖建力,张静.联合人脸与指纹的多模态生物特征识别方法综述[J].上海理工大学学报,2017,39(1):51-57.XIAO Jianli,ZHANG Jing.Survey on the multimodal biometric identification by fusing faces and fingerprints[J].Journal of University of Shanghai for Science and Technology,2017,39(1):51-57.
• [2]李艳萍,姜颖,胡金明,等.基于曲波变换和余弦测度的人脸识别方法[J].计算机科学,2016,43(5):294-297.LI Yanping,JIANG Ying,HU Jinming,et al.Face recognition method based on curvelet transform and cosine rules[J].Computer science,2016,43(5):294-297.
• [3]杨培德,黄仁裕,尤俊生.基于生物特征识别的终端安全管理系统的算法研究[J].计算机科学,2016,43(12):225-227.YANG Peide,HUANG Renyu,YOU Junsheng.Study on algorithms for terminal security management system based on biometrics recognition[J].Computer science,2016,43(12):225-227.
• [4]栗科峰,骆继明.基于局部结构分解的人脸图像特征提取方法[J].激光杂志,2015,36(11):71-74.LI Kefeng,LUO Jiming.Face feature extraction method based on local structure image decomposition[J].Laser journal,2015,36(11):71-74.
• [5]吴从中,刘渠芬,詹曙.基于稀疏表达的人脸遮挡物去除[J].合肥工业大学学报(自然科学版),2015(5):615-617.WU Congzhong,LIU Qufen,ZHAN Shu.A face occlusion removal method based on sparse representation[J].Journal of Hefei University of Technology(Natural science),2015(5):615-617.
• [6]唐守军.基于人脸识别和密码认证的文件夹加密方法[J].现代电子技术,2016,39(3):80-83.TANG Shoujun.Method of folder encryption based on face recognition and password authentication[J].Modern electronics technique,2016,39(3):80-83.
• [7]韩涛.基于比率图的表情映射机器视觉[J].科学技术与工程,2017,17(14):239-242.HAN Tao.Expressive expression mapping with ratio images of machine vision[J].Science technology and engineering,2017,17(14):239-242.
• [8]杨巨成,刘娜,房珊珊,等.基于深度学习的人脸识别方法研究综述[J].天津科技大学学报,2016,31(6):1-10.YANG Jucheng,LIU Na,FANG Shanshan,et al.Review of face recognition methods based on deep learning[J].Journal of Tianjing University of Science&Technology,2016,31(6):1-10.
• [9]马晓,张番栋,封举富.基于深度学习特征的稀疏表示的人脸识别方法[J].智能系统学报,2016,11(3):279-286.MA Xiao,ZHANG Fandong,FENG Jufu.Sparse representation via deep learning features based face recognition method[J].CAAI transactions on intelligent systems,2016,11(3):279-286.
• [10]赵喜玲.步态与人脸融合的身份识别方法[J].信阳师范学院学报(自然科学版),2017,30(3):460-463.ZHAO Xiling.Research on human identification based on gait and face[J].Journal of Xinyang Normal University(Natural science edition),2017,30(3):460-463.
• [11]李明夏,陈斌.基于监控视频的稳定实时人脸跟踪系统构建[J].计算机应用,2015(z2):250-253.LI Mingxia,CHEN Bin.Robust and real-time face tracking system construction based on surveillance video[J].Journal of computer applications,2015(S2):250-253.