基于子模式行列方向二维线性判别分析特征融合的特征提取

计算机应用 ›› 2014, Vol. 34 ›› Issue (12): 3593-3598.

基于子模式行列方向二维线性判别分析特征融合的特征提取

董晓庆,陈洪财

韩山师范学院物理与电子工程系,广东潮州 521041

收稿日期:2014-07-09 修回日期:2014-08-21 出版日期:2014-12-01 发布日期:2014-12-31
通讯作者: 董晓庆
作者简介:董晓庆(1982-）,男,广东潮州人,讲师,硕士,主要研究方向:图像处理、模式识别、自动控制；陈洪财(1967-）,男,山东菏泽人,副教授,硕士,主要研究方向:智能控制、嵌入式系统。
基金资助:
国际科技合作项目;2013年教育部高等学校“专业综合改革试点”项目;2013年韩山师范学院青年基金资助项目;2014年广东省高等教育教学改革项目

Feature extraction using a fusion method based on sub-pattern row-column two-dimensional linear discriminant analysis

DONG Xiaoqing,CHEN Hongcai

Department of Physics and Electronic Engineering, Hanshan Normal University, Chaozhou Guangdong 521041,China

Received:2014-07-09 Revised:2014-08-21 Online:2014-12-01 Published:2014-12-31
Contact: DONG Xiaoqing

摘要/Abstract

摘要：

针对人脸识别中表情和光照变化引起的面部变化、灰度不均匀等识别问题,提出一种基于子模式行列方向二维线性判别分析(Sp-RC2DLDA)的特征提取方法。该方法通过对原图像进行子模式分块处理,能有效提取图像的局部特征,减少表情、光照变化的影响,通过把相同位置的子图像组成子样本集,合理利用了子块间的空间关系,进一步提高了识别率；同时,对各个子样本集分别利用行方向二维线性判别分析(2DLDA)和列方向扩展2DLDA(E2DLDA)进行特征抽取,得到互补的行、列方向子图像特征,并分别把子图像特征组合成原图像的特征矩阵,然后利用一种特征融合方法对行、列方向特征矩阵进行有效融合,对互补的特征空间进行融合有效地改善了识别性能；最后采用最近邻分类器进行人脸识别实验。在Yale及ORL人脸库上的实验结果表明,Sp-RC2DLDA有效地减少了表情和光照变化的影响,具有较好的鲁棒性。

Abstract:

In order to solve the problems, such as facial change and uneven gray, caused by the variations of expression and illumination in face recognition, a novel feature extraction method based on Sub-pattern Row-Column Two-Dimensional Linear Discriminant Analysis (Sp-RC2DLDA) was proposed. In the proposed method, by dividing the original images into smaller sub-images, the local features could be extracted effectively, and the impact of variations in facial expression and illumination was reduced. Also, by combining the sub-images at the same position as a subset, the recognition performance could be improved for making full use of the spatial relationship among sub-images. At the same time, two classes of features which complemented each other can be obtained by synthesizing the local sub-features which were achieved by performing 2DLDA (Two-Dimensional Linear Discriminant Analysis) and Extend 2DLDA (E2DLDA) on a set of partitioned sub-patterns in the row and column directions, respectively. Then, the recognition performance was expected to be improved by employing a fusion method to effectively fuse these two classes of complementary features. Finally, nearest neighbor classifier was applied for classification. The experimental results on Yale and ORL face databases show that the proposed Sp-RC2DLDA method reduces the influence of variations in illumination and facial expression effectively, and has better robustness and classification performance than the other related methods.

中图分类号:

TP391

董晓庆陈洪财. 基于子模式行列方向二维线性判别分析特征融合的特征提取[J]. 计算机应用, 2014, 34(12): 3593-3598.

DONG Xiaoqing CHEN Hongcai. Feature extraction using a fusion method based on sub-pattern row-column two-dimensional linear discriminant analysis[J]. Journal of Computer Applications, 2014, 34(12): 3593-3598.

参考文献

［1］ADINI Y, MOSES Y, ULLMAN S. Face recognition: the problem of compensating for changes in illumination direction［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1997, 19(7): 721-732.
［2］BELHUMEUR P N, HESPANHA J P, KRIEGMAN D J. Eigenfaces vs. Fisherfaces: recognition using class specific linear projection［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1997,19(7):711-720.
［3］ZHOU D, YANG X, PENG N. A modified linear discriminant analysis and its application to face recognition［J］.Journal of Shanghai Jiaotong University,2005,39(4):527-530.(周大可, 杨新, 彭宁嵩. 改进的线性判别分析算法及其在人脸识别中的应用［J］.上海交通大学学报, 2005, 39(4):527-530.)
［4］ZHUANG H, ZHANG A, LI F. Based on an optimized LDA algorithm for face recognition［J］. Journal of Electronics and Information Technology, 2007, 29(9):2047-2049.
［5］YU H, YANG J. A direct LDS algorithm for high-dimensional data with application to face recognition［J］. Pattern Recognition, 2001, 34(10): 2067-2070.
［6］YANG J, ZHANG D, FRANGI A F, et al. Two-dimensional PCA: a new approach to apperarance-based face representation and recognition［J］.IEEE Transactions on Pattern Analysis and Machine Intelligence, 2004, 26(1):131-137.
［7］LI M, YUAN B. 2D-LDA: a statistical linear discriminant analysis for image matrix［J］. Pattern Recognition Letter, 2005,26(5):527-532.
［8］YANG J, ZHANG D, XU Y, et al. Two-dimensional discriminant transform for face recognition［J］. Pattern Recognition, 2005,38(7):1125-1129.
［9］JIN Y, RUAN Q. Face recognition using assembled matrix distance metric based 2DLDA algorithm［C］// Proceedings of the 8th International Conference on Signal Processing. Piscataway: IEEE, 2006:16-20.
［10］NOUSHATH S, KUMAR G H, SHIVAKUMARA P.(2D)2LDA: an efficient approach for face recognition［J］. Pattern Recognition, 2006,39(7):1396-1400.
［11］DU H,CHAI X,WANG F, et al. Face recognition using a fusion method based on bidirectional 2DLDA［J］.Chinese Journal of Scientific Instrument, 2009, 30(9):1880-1885.(杜海顺, 柴秀丽, 汪凤泉, 等. 一种基于双向 2DLDA 特征融合的人脸识别方法［J］. 仪器仪表学报, 2009, 30(9): 1880-1885.)
［12］WANG W, YAN Y. LDA for face recognition based on images segmentation［J］.Computer Engineering and Design, 2007, 28(12): 2889-2891.(王文豪, 严云洋. 基于图像分块的 LDA 人脸识别［J］. 计算机工程与设计, 2007, 28(12): 2889-2891.)
［13］YAN W, PENG G. Image feature extracting approach based on blocked FLD［J］. Science Technology and Engineering, 2006, 6(19):3107-3110.(延伟东, 彭国华. 基于分块 FLD 的图像特征提取方法［J］. 科学技术与工程, 2006, 6(19):3107-3110.)
［14］WANG L,WU J, JIA L. Face recognition method based on bidirectional and modular 2DLDA［J］.Journal of Electronic Measurement and Instrument, 2013, 27(8): 760-765.(王磊, 武敬飞, 贾莉. 一种基于双向模块 2DLDA 的人脸识别方法［J］. 电子测量与仪器学报, 2013, 27(8): 760-765.)
［15］CHEN S, ZHU Y. Subpattern-based principal component analysis［J］. Pattern Recognition, 2004, 37(1):1081-1083.
［16］HAN K, ZHU X. Research on two-dimensional LDA for face recognition［J］. Journal of Electronics: China, 2006, 23(6): 943-947.
［17］LI M, YUAN B. 2D-LDA: a statistical linear discriminant analysis for image matrix［J］. Pattern Recognition Letters, 2005, 26(5):527-532.

[1]	吴军欧阳艾嘉张琳. 基于影响度的统计显著序列模式挖掘算法[J]. 计算机应用, 0, (): 0-0.
[2]	张璐方春祝铭. 基于Res2Net-YOLACT和融合特征的室内跌倒检测算法[J]. 计算机应用, 0, (): 0-0.
[3]	殷雨昌王洪元陈莉冯尊登肖宇. 基于单标注样本的多损失学习与联合度量视频行人重识别[J]. 计算机应用, 0, (): 0-0.
[4]	胡军许正康刘立钟福金张清华. 融合多粒度社区信息的网络嵌入方法[J]. 计算机应用, 0, (): 0-0.
[5]	李润泽孙雪姣. 基于时间条件提取序列的数据流偏好查询[J]. 计算机应用, 0, (): 0-0.
[6]	罗圣钦陈金怡李洪均. 基于注意力机制的多尺度残差UNet实现乳腺癌灶分割[J]. 计算机应用, 0, (): 0-0.
[7]	曹一珉蔡磊高敬阳. 基于生成对抗网络的基因数据生成方法[J]. 计算机应用, 0, (): 0-0.
[8]	陈冲闫珠赵继轩何为梁华庆. 基于集合经验模态分解和长短期记忆网络的催化裂化装置NOx排放预测[J]. 计算机应用, 0, (): 0-0.
[9]	徐光柱林文杰陈莎匡婉雷帮军周军. U-Net与自适应阈值脉冲耦合神经网络相结合的眼底血管分割方法[J]. 计算机应用, 0, (): 0-0.
[10]	杨鼎康黄帅王顺利翟鹏李一丹张立华. 基于对抗生成网络和网络集成的面部表情识别方法EE-GAN[J]. 计算机应用, 0, (): 0-0.
[11]	李讷徐光柱雷帮军马国亮石勇涛. 交通道路行驶车辆车标识别算法[J]. 计算机应用, 0, (): 0-0.
[12]	孟杰王莉杨延杰廉飚. 基于多模态深度融合的虚假信息检测[J]. 计算机应用, 0, (): 0-0.
[13]	秦庭威赵鹏程秦品乐曾建朝柴锐黄永琦. 基于残差注意力机制的点云配准算法[J]. 计算机应用, 0, (): 0-0.
[14]	鲁永帅唐英杰马鑫然. 基于深度特征融合的无纺布低对比度浆丝缺陷检测方法[J]. 计算机应用, 0, (): 0-0.
[15]	王宇航周永霞吴良武. 基于高斯函数的池化算法[J]. 计算机应用, 0, (): 0-0.