«上一篇/Previous Article|本期目录/Table of Contents|下一篇/Next Article»

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:

32

期数:

2022年01期

页码:

104-110

栏目:

图形与图像

出版日期:

2022-01-10

文章信息/Info

Title:

Face Recognition with a Single Training Sample Per Person Based onKernel Extended Hybrid Block Dictionary Learning

文章编号:

1673-629X(2022)01-0104-07

作者:

马杲东1 ; 吕 非2 ; 童 莹3 ; 曹雪虹3

1. 南京邮电大学 通信与信息工程学院,江苏 南京 210003;

2. 南瑞集团有限公司,江苏 南京 211106;

3. 南京工程学院 信息与通信工程学院,江苏 南京 211167

Author(s):

MA Gao-dong1 ; LYU Fei2 ; TONG Ying3 ; CAO Xue-hong3

1. School of Telecommunications and Information Engineering,Nanjing University ofPosts and Telecommunications,Nanjing 210003,China;

2. NARI Group Corporation,Nanjing 211106,China;

3. School of Information and Communication Engineering,Nanjing Institute o

关键词:

稀疏表示分类; 核判别分析; 人脸识别; 混合块字典; 单样本

Keywords:

sparse representation - based classification;  kernel discriminant analysis;  face recognition;  hybrid block dictionary;  singlesample per person

分类号:

TP273

DOI:

10. 3969 / j. issn. 1673-629X. 2022. 01. 018

摘要:

稀疏表示分类(sparse representation-based classification,SRC) 在样本数量充足下的人脸识别中具有较好的识别效果。 然而由于基本字典缺乏判别性同时过度依赖于字典中每类样本的原子数目,稀疏表示分类在真实情况下的单样本(每类样本只有一张训练样本) 人脸识别任务中缺乏鲁棒性。 针对以上问题,该文提出了基于核扩展混合块字典的单样本人脸识别方法。 首先,对样本进行分块处理,分别对分块图像进行核判别分析( kernel discriminant analysis,KDA) 投影降维,提取图像的局部特征信息构成更具判别性的基本块字典;然后,为经过 KDA 投影之后的分块样本分别构建遮挡字典和类内差异字典来描述样本中的大面积连续遮挡以及光照、表情等类内差异信息,将遮挡字典和类内差异字典共同组合成混合块字典,使混合块字典能够更好地描述测试样本中不同类型的差异信息;最后,将测试样本表示为基本块字典和混合块字典的稀疏线性组合,根据重构残差进行分类识别,从而实现真实情况下的单样本人脸识别。 在标准人脸库 CAS-PEAL,AR 以及真实人脸库 LFW 和 PubFig 上的实验结果表明,该方法与其他方法相比有较好的结果。

Abstract:

Sparse representation-based classification ( SRC) is effective in face recognition with sufficient samples. However,due to thelack of discriminativeness of its basic dictionary and excessive dependence on the number of atoms of each class of sample in thedictionary,SRC lacks robustness in face recognition tasks of single sample per person ( SSPP) . Therefore,we propose a single sampleface recognition method based on kernel extended hybrid block dictionary. Firstly,the samples are divided into blocks,and the kernel discriminant analysis ( KDA ) projection dimension reduction is performed on the divided images respectively, and the local featureinformation of the image is extracted to form a more discriminative basic block dictionary. Then,for the blocked samples after KDA projection,an occlusion dictionary and an intra-class difference dictionary are constructed to describe the large-area continuous occlusion inthe sample,as well as the intra-class difference information such as illumination and expression. The occlusion dictionary and the intra-class dictionary are combined to construct the hybrid block dictionary and enable the hybrid block dictionary to better describe the intra-class variation in the test sample. Finally,the test samples are represented as sparse linear combinations of the basic block dictionary andthe hybrid block dictionary. The classification is determined by the reconstruction residuals. Experiment on standard face databases CAS-PEAL,AR and real face databases LFW and PubFig shows that the proposed method has better results compared with other methods.

常用功能

更新日期/Last Update: 2022-01-10