An Improved Two-Step Face Recognition Algorithm Based on Sparse Representation

Yongjun Zhang,Qian Wang,Zhongwei Cui,Ling Xiao

doi:10.1109/access.2019.2940876

Yongjun Zhang, Qian Wang + Show 2 more

Open Access

https://doi.org/10.1109/access.2019.2940876

Copy DOI

Journal: IEEE Access	Publication Date: Jan 1, 2019
Citations: 5	License type: CC BY 4.0

Affiliation: Guizhou University, Guizhou Education University

Abstract

Weighted score fusion is a widely used score fusion scheme, but the weights need to be set manually. The results generally vary greatly when the weights are different, so it is difficult to find the optimal weights. This is why it is necessary to constantly set different weights for experimental comparisons to find the optimal weights. In this paper, an improved fusion method is proposed for above shortcoming, that is, multiplication fusion applicable to sparse representation. The fusion scheme not only is easy to use but also does not need to be artificially set weights. Moreover, it is consistent with the correlation between the classification error and the score obtained by the experimental analysis. In the field of face recognition, it has been shown that the two-step face recognition (TSFR) based on representation using the original training samples and the generated “symmetric face” training samples can achieve excellent face recognition performance. Face recognition based on multiplication fusion and TSFR proposed in this paper can further improve the recognition accuracy.

Highlights

Face recognition is a kind of biometric recognition technology based on human facial feature information, so the field of face recognition has been concerned by people [1], [3]–[5]
Training samples; the second step and the third step respectively use the original training samples and the ‘‘symmetric face’’ samples to perform two-step face recognition; the last step is the weighted score fusion of the original algorithm is transformed into multiplication fusion by using the scores obtained in the second and third steps
The improved multiplication fusion method is used for score fusion to reduce the final classification error rates

Summary

INTRODUCTION

Face recognition is a kind of biometric recognition technology based on human facial feature information, so the field of face recognition has been concerned by people [1], [3]–[5]. Rahimzadeh Arashloo et al [19] proposed a new method for single sample face recognition problem based on local dual-tree complex wavelet transform (DT-CWT) representation which offers invariance to moderate real world image variations, such as illumination, expression, head pose, shift and in-plane rotation. Feng et al [21] presented a novel face recognition method based on direct discriminant Volterra kernels and effective feature classification (DD-VK). They presented an efficient robust method, namely R1 -2-DPCA for feature extraction which is robust to outliers and helps encode discriminant information These classic algorithms have made a huge contribution to the accurate recognition of faces. Training samples; the second step and the third step respectively use the original training samples and the ‘‘symmetric face’’ samples to perform two-step face recognition; the last step is the weighted score fusion of the original algorithm is transformed into multiplication fusion by using the scores obtained in the second and third steps

ALGORITHM INTRODUCTION

IMPROVED ALGORITHM

CONCLUSION