基于多特征融合的人脸识别算法

doi:10.16180/j.cnki.issn1007-7820.2019.07.009

电子科技 ›› 2019, Vol. 32 ›› Issue (7): 43-48.doi: 10.16180/j.cnki.issn1007-7820.2019.07.009

基于多特征融合的人脸识别算法

苏饶,李菲菲,陈虬

上海理工大学光电信息与计算机工程学院,上海 200093

收稿日期:2018-07-16 出版日期:2019-07-15 发布日期:2019-08-14
作者简介:苏饶(1993-),女,硕士研究生。研究方向:计算机视觉与模式识别。|李菲菲(1970-),女,博士,教授。研究方向:多媒体信息处理、图像处理与模式识别、信息检索等。|陈虬(1972-),男,博士,教授。研究方向:图像处理与模式识别、计算机视觉、信息检索等。
基金资助:
上海市高校特聘教授(东方学者) 岗位计划(ES2015XX)

Face Recognition Algorithm Based on Multiple Feature Fusion

SU Rao,LI Feifei,CHEN Qiu

School of Optical Electrical and Computer Engineering,University of Shanghai for Science and Technology,Shanghai 200093,China

Received:2018-07-16 Online:2019-07-15 Published:2019-08-14
Supported by:
The Program for Professor of Special Appointment (Eastern Scholar) at Shanghai Institutions of Higher Learning(ES2015XX)

摘要/Abstract

摘要：

针对局部二值模式描述子提取的纹理信息以及梯度幅值量化算子提取的边缘特征无法有效且全面地描述人脸信息的问题,文中提出一种基于马尔可夫稳态特征模型的多特征融合算法。首先,将通过GMQ算子提取的边缘特征以及通过LBP描述子提取的纹理特征分别与马尔可夫稳态特征模型进行融合,然后再将两者进行有效地线性加权融合。最后,在ORL数据集上进行的实验显示,文中提出算法的识别精度可达到 95.83%。与单一的特征提取算法以及常见的人脸识别算法对比结果表明了该方法的有效性。

关键词: 人脸识别, 局部二值模式, 梯度幅值量化, 马尔可夫稳态特征, 线性加权融合, ORL数据集

Abstract:

The texture information extracted by the LBP descriptor and the edge feature extracted by the GMQ operator cannot effectively and comprehensively describe the facial feature. To solve the problems, a novel multiple feature fusion algorithm based on Markov Stationary Features model was proposed. Firstly, the edge features obtained by GMQ operator as well as the texture features by LBP descriptor were fused with the MSF model respectively. Then the two MSF-based features were fused by linear weighting. Finally, experiments on the ORL dataset showed that the proposed algorithm could achieve an accuracy of 95.83%. Compared with a single feature extraction algorithm and a common face recognition algorithm, the effectiveness of the proposed method was proved.

Key words: face recognition, local binary pattern, gradient magnitude quantization, markov stationary features, linear weighting fusion, ORL database

中图分类号:

TP391.41

苏饶,李菲菲,陈虬. 基于多特征融合的人脸识别算法[J]. 电子科技, 2019, 32(7): 43-48.

SU Rao,LI Feifei,CHEN Qiu. Face Recognition Algorithm Based on Multiple Feature Fusion[J]. Electronic Science and Technology, 2019, 32(7): 43-48.

图/表 10

图1

图2

图3

图4

图5

图6

表1

表2

表3

表4

参考文献 22

[1]	Bartlett M S, Movellan J R, Sejnowski T J . Face recognition by independent component analysis[J]. IEEE Transactions on Neural Networks, 2002,13(6):1450-64.
[2]	Hyvärinen A . Fast and robust fixed-point algorithms for independent component analysis[J]. IEEE Transactions on Neural Networks, 1999,10(3):626.
[3]	Belhumeur P N, Kriegman D J . Eigenfaces vs. Fisherfaces: recognition using class specific linear projection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002,19(7):711-720.
[4]	Lowe D G . Distinctive image features from scale-invariant keypoints[J]. International Journal of Computer Vision, 2004,60(2):91-110. doi: 10.1023/B:VISI.0000029664.99615.94
[5]	ulina J K J,Sharmila T S., . Facial recognition using histogram of gradients and support vector machines [C].Chengnai:International Conference on Computer, Communication and Signal Processing ( ICCCSP), 2017.
[6]	Pietikäinen M, Ojala T, Xu Z . Rotation-invariant texture classification using feature distributions[J]. Pattern Recognition, 2000,33(1):43-52.
[7]	杨恢先, 翟云龙, 蔡勇勇 , 等. 基于中心对称梯度幅值相位模式的单样本人脸识别[J].光电子·激光, 2015(5):969-977.
	Yang Huixian, Zhai Yunlong, Cai Yongyong , et al. Face recognition based on center-symmetric gradient magnitude and phase pattern[J]. Journal of Optoelectronics·Laser, 2015(5):969-977.
[8]	杨利平, 辜小花 . 用于人脸识别的相对梯度直方图特征描述[J]. 光学精密工程, 2014,22(1):152-159.
	Yang Liping, Gu Xiaohua . Relative gradient histogram features for face recognition[J]. Optics and Precision Engineering, 2014,22(1):152-159.
[9]	Zhu C . Local multiple patterns based multiresolution gray-scale and rotation invariant texture classification[M]. New York:Elsevier Science Inc, 2012.
[10]	Borade S N, Deshmukh P R, Ramu S . Face recognition using fusion of PCA and LDA:borda count approach[C].Athens:Mediterranean Conference on Control and Automation ( MED), 2016.
[11]	焦淑红, 孙志帅 . 基于GPCA的KNNY与SVM融合的人脸识别方法[J]. 电子科技, 2016,29(2):74-76.
	Jiao Shuhong, Sun Zhishuai . Face recognition based of GPCA of KNN and SVM fusion[J]. Electronic Science and Technology, 2016,29(2):74-76.
[12]	Chen Q, Kotani K, Lee F , et al. Face recognition using markov stationary features and vector quantization histogram[C].Chengdu:International Conference on Computational Science and Engineering( CSE), 2014.
[13]	Lee F, Kotani K, Chen Q , et al. A robust face recognition algorithm using Markov stationary features and adjacent pixel intensity difference quantization histogram[J]. International Journal of Advancements in Computing Technology, 2011,4(12):335-343.
[14]	Wang A, Xu Y . Markov chain analysis and prediction of consumer price index in Xinjiang[C].Changsha:International Conference on Measuring Technology and Mechatronics Automation ( ICMTMA), 2018.
[15]	Li J, Wu W, Wang T, et al. One step beyond histograms: Image representation using Markov stationary features [C].Anchorage:IEEE Conference on Computer Vision and Pattern Recognition, 2008.
[16]	Hamid Sadeghi, Abolghasem-A Raie . Approximated Chi-Square distance for histogram matching in facial image analysis: face and expression recognition[C].Isfahan:Iranian Conference on Machine Vision and Image Processing ( MVIP), 2017.
[17]	黄华盛, 杨阿庆 . 基于PCA算法的人脸识别[J]. 电子科技, 2015,28(8):98-101.
	Huang Huasheng, Yang Aqing . Face recognition on PCA algorithm[J]. Electronic Science and Technology, 2015,28(8):98-101.
[18]	Ren J, Jiang X, Yuan J . LBP encoding schemes jointly utilizing the information of current bit and other LBP bits[J]. IEEE Signal Processing Letters, 2015,22(12):2373-2377.
[19]	Xu J Q, Li X P, Dong H J, et al. Face recognition width improved local binary patterns [C].Hong Kong: International Conference on Intelligent Computation and Industrial Application, 2011.
[20]	Du M, Chellappa R . Face association for videos using conditional random fields and Max-Margin Markov networks[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2016,38(9):1762-1773.
[21]	Chen Q, Kotani K, Lee F . Face recognition using multiple histogram features in spatial and frequency domains[C].Naples:International Conference on Signal-Image Technology & Internet-Based Systems ( SITIS), 2016.
[22]	Rajesh K M, Naveen Kumar M . A robust method for face recognition and face emotion detection system using support vector machines[C]. Mysuru:International Conference on Electrical, Electronics,Communication, Computer and Optimization Techniques ( ICEECCOT), 2016.

滤波器模板尺寸	识别率
无	79.17%
3×3	86.67%
5×5	90.00%
7×7	89.17%
9×9	85.00%
11×11	6.67%
13×13	84.17%
15×15	84.17%
17×17	83.30%

等价模式	识别率
无	75.00%
58种	86.67%

ω ₁	识别率
1	90.00%
1/10	90.00%
2/10	91.67%
3/10	95.83%
4/10	95.83%
5/10	94.17%
6/10	93.30%
7/10	93.30%
8/10	93.30%
9/10	90.00%
0	86.67%

算法	识别率
LBP	84.10%
ICA	85.00%
PCA	85.30%
HOG	90.24%
MSF-LBP	86.67%
MSF-GMQ	90.00%
GMQ-MSF-LBP	95.83%

基于多特征融合的人脸识别算法

Face Recognition Algorithm Based on Multiple Feature Fusion

RichHTML

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

图/表 10

参考文献 22

相关文章 15

Metrics

本文评价

推荐阅读 3

[1]	程晓雅,张雷. 基于深度学习的有遮挡人脸识别方法研究[J]. 电子科技, 2022, 35(1): 35-39.
[2]	彭荣杰,彭亚雄,陆安江. 基于改进PCA+SVM的人脸识别系统[J]. 电子科技, 2021, 34(12): 56-61.
[3]	王春江,张猛,张建飞. 基于网络摄像头的活体人脸识别系统设计[J]. 电子科技, 2020, 33(6): 63-68.
[4]	司琴,李菲菲,陈虬. 基于深度学习与特征融合的人脸识别算法[J]. 电子科技, 2020, 33(4): 18-22.
[5]	刘芬. 基于线性和非线性算法的高效人脸识别系统[J]. 电子科技, 2019, 32(7): 82-86.
[6]	胡少聪. 基于深度学习的人脸识别方法研究[J]. 电子科技, 2019, 32(6): 82-86.
[7]	周柱,甘屹,孙福佳. 一种融合Gabor+SIFT特征的人脸识别算法[J]. 电子科技, 2019, 32(4): 1-5.
[8]	焦淑红,孙志帅. 基于GPCA的KNNY与SVM融合的人脸识别方法[J]. , 2016, 29(2): 74-.
[9]	黄华盛,杨阿庆. 基于PCA算法的人脸识别[J]. , 2015, 28(8): 98-.
[10]	张洋洋,张会林,陈茉莉. 基于二代身份证照的多姿态人脸图像生成[J]. , 2015, 28(6): 64-.
[11]	穆新亮. 基于混合核函数的快速KPCA人脸识别算法[J]. , 2015, 28(2): 46-.
[12]	郑颖. 一种快速非负矩阵分解的人脸识别算法[J]. , 2015, 28(2): 51-.
[13]	杨阿庆,黄华盛. 基于Fihserfaces算法的人脸识别[J]. , 2015, 28(10): 11-.
[14]	穆新亮, 武亚静. 基于双向主成分分析和压缩感知的人脸识别算法[J]. , 2014, 27(1): 45-.
[15]	王宏勇,常燕,王青青. 面向识别的人脸去光照预处理算法[J]. , 2013, 26(3): 153-.