首页 > 论文 > 液晶与显示 > 32卷 > 12期(pp:987-992)

混合样本协同表示算法的人脸识别研究

Face recognition research based on variant samples and collaborative representation

  • 摘要
  • 论文信息
  • 参考文献
  • 被引情况
  • PDF全文
分享:

摘要

在人脸识别中, 人脸图像受到表情、光照、遮挡、姿态变化、特别是训练样本数量的影响, 而现实中经常只获得少量的训练样本, 由于原始样本生成虚拟样本可以增加训练样本的数量, 分析提出原始样本与轴对称样本融合的协同表示算法。首先生成镜像样本与轴对称样本, 再在协同表示分类器下分类, 最后加权值融合, 分析不同权值下的人脸识别率。实验结果显示原始样本、镜像样本与轴对称样本融合能提高识别率, 而原始样本与轴对称样本融合的识别率更加优越, 较原始样本, 识别率提高2%~9%, 比原始样本与镜像样本融合高1%~5%。结果表明本文提出方法能有效提高人脸识别率。

Abstract

For face recognition, the face image are affected by the variations of expression, lighting, occlusion, pose and especially the number of training samples. However, in practical application, we only have insufficient training samples. The collaborative representation algorithm of combining the original training samples with the axial-symmetry samples is proposed because the original training samples generate the corresponding virtual training samples to increase the number of training samples. Firstly, the original training samples generate the corresponding mirror samples and axial-symmetry samples. Secondly, the reconstruction errors are obtained by using collaborative representation based classification. Finally, the variant reconstruction errors are combined with different weighted number to compare face recognition rates. The experimental results show that the face recognition rates are increased by combining the original training samples with the mirror samples and the axial-symmetry samples. The face recognition rates of combining the original training samples with the axialsy-mmetry samples are 2%~9% and 1%~5% better than the original training samples and the original training samples with the mirror samples respectively. It shows that the paper’s method is effective in face recognition.

投稿润色
补充资料

中图分类号:TP391

DOI:10.3788/yjyxs20173212.0987

所属栏目:图像处理

基金项目:四川理工学院科研项目(No.2015RC16)

收稿日期:2017-06-28

修改稿日期:2017-09-27

网络出版日期:--

作者单位    点击查看

杨明中:四川理工学院 自动化与信息工程学院,四川 自贡 643000
杨平先:四川理工学院 自动化与信息工程学院,四川 自贡 643000
林国军:四川理工学院 自动化与信息工程学院,四川 自贡 643000

联系人作者:杨明中(erdenzer@qq.com)

备注:杨明中(1991-), 男, 四川南充人, 硕士研究生, 主研方向为图像处理、模式识别。

【1】ZHU N B, LI S T. A kernel-based sparse representation method for face recognition [J]. Neural Computing and Applications, 2014, 24(3): 845-852.

【2】DU Y, WANG Y. Generating virtual training samples for sparse representation of face images and face recognition [J].Journal of Modern Optics, 2016, 63(6): 536-544.

【3】闵卫东, 石杰, 韩清, 等.一种分布式人脸识别方法及性能优化[J].光学 精密工程, 2017, 25(3): 779-785.
MIN W D, SHI J, HAN Q,et al. A distributed face recognition method and performance optimization [J]. Optics and Precision Engineering, 2017, 25(3): 779-785. (in Chinese)

【4】WU J X, ZHOU Z H. Face recognition with one training image per person [J]. Pattern Recognition Letters, 2002, 23(14): 1711-1719.

【5】XU Y, ZHU X J, LI Z M, et al. Using the original and ‘symmetrical face’ training samples to perform representation based two-step face recognition [J]. Pattern Recognition, 2013, 46(4): 1151-1158.

【6】XU Y, LI X L, YANG J, et al. Integrate the original face image and its mirror image for face recognition [J]. Neurocomputing, 2014, 131: 191-199.

【7】TANG D Y, ZHU N B, FU Y, et al. A novel sparse representation method based on virtual samples for face recognition [J]. Neural Computing and Applications, 2014, 24(3/4): 513-519.

【8】WRIGHT J, YANG A Y, GANESH A, et al. Robust face recognition via sparse representation [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2009, 31(2): 210-227.

【9】ZHANG L, YANG M, FENG X C. Sparse representation or collaborative representation: which helps face recognition? [C]//Proceedings of 2011 International Conference on Computer Vision. Barcelona, Spain: IEEE, 2011: 471-478.

【10】魏冬梅, 周卫东.基于位平面和协作表示的人脸识别算法[J].北京理工大学学报, 2014, 34(9): 966-971.
WEI D M, ZHOU W D. Facerecognition using bit-plane images and collaborative representation [J]. Transactions of Beijing Institute of Technology, 2014, 34(9): 966-971. (in Chinese)

【11】ZHU N B, CHEN S X. Weighted sparse representation based on virtual test samples for face recognition [J].Optik-International Journal for Light and Electron Optics, 2017, 140(7): 853-859.

【12】项晓丽, 武圣, 龙伟, 等.融合原始样本和虚拟样本的人脸识别算法[J].电视技术, 2016, 40(9): 117-121.
XIANG X L, WU S, LONG W,et al. Face recognition based on the fusion of original sample and virtual sample [J]. Video Engineering, 2016, 40(9): 117-121. (in Chinese)

引用该论文

YANG Ming-zhong,YANG Ping-xian,LIN Guo-jun. Face recognition research based on variant samples and collaborative representation[J]. Chinese Journal of Liquid Crystals and Displays, 2017, 32(12): 987-992

杨明中,杨平先,林国军. 混合样本协同表示算法的人脸识别研究[J]. 液晶与显示, 2017, 32(12): 987-992

您的浏览器不支持PDF插件,请使用最新的(Chrome/Fire Fox等)浏览器.或者您还可以点击此处下载该论文PDF