基于局部强化最小二乘回归分类法的人脸识别方法

(厦门大学嘉庚学院信息科学与技术学院,福建 漳州 363105)

人脸识别; 最小二乘回归; 局部强化; 自适应; 分类

Face recognition using local strengthen least-square regression classification method
JIAN Cairen,XIA Jingbo*

(School of Information Science & Technology,Xiamen University Tan Kah kee College,Zhangzhou 363105,China)

face recognition; least square regression; local strengthen; adaptive; classification

DOI: 10.6043/j.issn.0438-0479.201809025

备注

为解决基于表示理论的分类法未考虑噪声样本对重构系数影响的不足,利用局部约束协同表示法改进最小二乘回归分类法,提出局部强化最小二乘回归分类法.该方法通过非负稀疏表示自适应选择近邻样本,并利用近邻样本的协同作用强化重构系数使得局部强化最小二乘回归分类法具有较好的鲁棒性和容噪性.该方法可以克服传统分类方法存在的过拟合问题.在4个人脸图像数据集上的实验结果表明该方法可以提高人脸识别准确率.

In order to improve the weakness of classification method based on the representation theorythat ignore the influence of noise on the reconstruction coefficients,we propose a local strengthen least-square regression classification method to improve the least-square regression classification method by using the local-constraint cooperative representation.The proposed method can select neighbor samples adaptively by using nonnegative sparse representation.It strengthens reconstruction coefficients by using neighbor data samples,and improves anti-noise ability.Furthermore,it can overcome the over-fitting problems that plague traditional classification methods.Experimental results on the four face recognition datasets show that this method can improve recognition accuracies.