结合小波与递归神经网络的低分辨率人脸识别

doi:10.19665/j.issn1001-2400.2019.06.014

西安电子科技大学学报 ›› 2019, Vol. 46 ›› Issue (6): 95-101.doi: 10.19665/j.issn1001-2400.2019.06.014

结合小波与递归神经网络的低分辨率人脸识别

欧阳宁,王先傲,蔡晓东(),林乐平

桂林电子科技大学信息与通信学院, 广西壮族自治区桂林 541004

收稿日期:2019-04-15 出版日期:2019-12-20 发布日期:2019-12-21
通讯作者: 蔡晓东
作者简介:欧阳宁(1972—),男,教授,E-mail:ynou@guet.edu.cn
基金资助:
国家自然科学基金(61661017);中国博士后科学基金(2016M602923XB);认知无线电与信息处理重点实验室基金(CRKL160104);认知无线电与信息处理重点实验室基金(CRKL150103);认知无线电与信息处理重点实验室基金(2011KF11);桂林科技开发项目(20150103-6);2018年新疆维吾尔自治区重点研发计划(2018B03022-1);2018年新疆维吾尔自治区重点研发计划(2018B03022-2);广西自然科学基金(2017GXNSFBA198212);广西自然科学基金(2014GXNSFDA118035);广西自然科学基金(2016GXNSFAA38014);桂林电子科技大学研究生教育创新计划(2016YJCXB02);广西科技创新能力与条件建设计划(桂科能1598025-21)

Low resolution face recognition method based on wavelet and recursive neural networks

OUYANG Ning,WANG Xian’ao,CAI Xiaodong(),LIN Leping

School of Information and Communication Engineering, Guilin University of Electronic Technology,Guilin 541004, China

Received:2019-04-15 Online:2019-12-20 Published:2019-12-21
Contact: Xiaodong CAI

摘要/Abstract

摘要：

针对低分辨率人脸图像缺少有效信息而导致识别率较低的问题,提出一种结合哈尔小波与递归神经网络的低分辨率人脸识别方法。首先,通过深层网络直接预测小波系数,经过小波逆变换得到高分辨率人脸图像,可以有效地重建高频信息;其次,在卷积神经网络中加入递归模块,在增加网络深度的同时减少参数冗余,提升模型的映射能力;最后,提出一种优化的重建与感知损失融合方法,将小波系数重建损失与感知损失进行加权融合,用以生成有利于识别的人脸图像。基于公开数据集,对图像重建质量与识别性能进行了对比。实验结果表明,即使在极低的分辨率条件下(8×8,16×16),仍然能够重建出更加锐利的人脸图像。在此基础上,其识别能力优于目前领先的超分辨率重建算法。

关键词: 哈尔小波, 递归神经网络, 人脸识别, 融合损失, 超分辨率重建

Abstract:

To improve the accuracy of low-resolution face recognition with limited information, a method based on the Haar wavelet and recurrent neural network is proposed. First, the wavelet coefficients are directly predicted through the deep neural network. High-resolution face images with high-frequency information can be reconstructed by the inverse wavelet transform. Second, a recursive module is added to the convolutional neural network to increase the depth of the network, which can reduce the redundancy of parameters effectively. Finally, a fusion loss method is utilized, in which the loss of wavelet coefficients reconstruction and the perceptual are weighted and fusioned to generate images for recognition. Based on open dataset, the image reconstruction quality and recognition performance are compared, respectively. Experimental results show that sharper face images can be reconstructed even with extremely low resolutions (8×8, 16×16), and that its recognition ability outperforms that of state-of-the-art face super resolution algorithms.

Key words: Haar wavelet, recursive neural network, face recognition, fusion loss, super resolution

中图分类号:

TP183

欧阳宁,王先傲,蔡晓东,林乐平. 结合小波与递归神经网络的低分辨率人脸识别[J]. 西安电子科技大学学报, 2019, 46(6): 95-101.

OUYANG Ning,WANG Xian’ao,CAI Xiaodong,LIN Leping. Low resolution face recognition method based on wavelet and recursive neural networks[J]. Journal of Xidian University, 2019, 46(6): 95-101.

图/表 10

图1

图2

表1

特征重建模块结构"

卷积模块	SphereFace^[7]
Conv1.x	$3 × 3,64 × 1, S 2 3 × 3,64 3 × 3,64 × 1$
Conv2.x	$3 × 3,128 × 1, S 2 3 × 3,128 3 × 3,128 × 2$
Conv3.x	$3 × 3,256 × 1, S 2 3 × 3,256 3 × 3,256 × 4$
Conv4.x	$3 × 3,512 × 1, S 2 3 × 3,512 3 × 3,512 × 1$
Fc	512

表1

图3

图4

表2

表3

表4

图5

表5

参考文献 17

[1]	DONGC, LOY C C, HE K , et al. Image Super-resolution Using Deep Convolutional Networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2016,38(2):295-307. doi: 10.1109/TPAMI.2015.2439281 pmid: 26761735
[2]	张瑞, 冯象初, 杨丽霞 , 等. 全局稀疏梯度耦合张量扩散的图像去噪模型[J]. 西安电子科技大学学报, 2017,44(6):150-155.
	ZHANG Rui, FENG Xiangchu, YANG Lixia , et al. Global Sparse Gradient Coupled Tensor Diffusion Model for Image Denoising[J]. Journal of Xidian University, 2017,44(6):150-155.
[3]	LI J, FANG F, MEI K , et al. Multi-scale Residual Network for Image Super-resolution[C]//Lecture Notes in Computer Science: 11212. Heidelberg: Springer Verlag, 2018: 527-542.
[4]	王世平, 毕笃彦, 刘坤 , 等. 一种多映射卷积神经网络的超分辨率重建算法[J]. 西安电子科技大学学报, 2018,45(4):155-160.
	WANG Shiping, BI Duyan, LIU Kun , et al. Multi-mapping Convolution Neural Network for the Image Super-resolution Algorithm[J]. Journal of Xidian University, 2018,45(4):155-160.
[5]	TAI Y, YANG J, LIU X . Image Super-resolution via Deep Recursive Residual Network[C]//Proceedings of the 2017 30th IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2017: 2790-2798.
[6]	HUANG H, HE R, SUN Z , et al. Wavelet-SRNet: a Wavelet-based CNN for Multi-scale Face Super Resolution[C]// Proceedings of the 2017 IEEE International Conference on Computer Vision. Piscataway: IEEE, 2017: 1698-1706.
[7]	JOHNSON J, ALAHI A, FEI-FEI L . Perceptual Losses for Real-time Style Transfer and Super-resolution[C]//Lecture Notes in Computer Science: 9906. Heidelberg: Springer Verlag, 2016: 694-711.
[8]	LIU W, WEN Y, YUZ Y . et al. Sphereface: Deep Hypersphere Embedding for Face Recognition[C]//Proceedings of the 2017 30th IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2017: 6738-6746.
[9]	IOFFE S, SZEGEDY C . Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift[C]//Proceedings of the 2015 32nd International Conference on Machine Learning. Lille: International Machine Learning Society, 2015: 448-456.
[10]	SIMONYAN K, ZISSERMAN A . Very Deep Convolutional Networks for Large-scale Image Recognition[C]// Proceedings of the 2015 3rd International Conference on Learning Representations. San Diego: International Conference on Learning Representations, 2015: 1556.
[11]	YI D, LEI Z, LIAO S C , et al. Learning Face Representation from Scratch[J/OL]. [2019-01-02]. https://arxiv.org/pdf/1411.7923.pdf.
[12]	HUANG G B, MATTAR M, BERG T , et al. Labeled Faces in the Wild: a Database for Studying Face Recognition in Unconstrained Environments [EB/OL]. [2019-03-26]. https://www.docin.com/p-1204016448.html.
[13]	ZHANG K, ZHANGZ, LI Z , et al. Joint Face Detection and Alignment Using Multi-task Cascaded Convolutional Networks[J]. IEEE Signal Processing Letters, 2016,23(10):1499-1503. doi: 10.1109/LSP.2016.2603342
[14]	KINGMA D P, BA J . Adam: a Method for Stochastic Optimization[J/OL].[2019-01-02]. https://arxiv.org/pdf/1412.6980.pdf.
[15]	MARCINIAK T, CHMIELEWSKA A, WEYCHAN R , et al. Influence of Low Resolution of Images on Reliability of Face Detection and Recognition[J]. Multimedia Tools and Applications, 2015,74(12):4329-4349. doi: 10.1007/s11042-013-1568-8
[16]	CHEN S, LIU Y, GAO X , et al. Mobilefacenets: Efficient CNNs for Accurate Real-time Face Verification on Mobile Devices[C]//Lecture Notes in Computer Science: 10996. Heidelberg: Springer Verlag, 2018: 428-438.
[17]	WEN Y, ZHANG K, LI Z , et al. A Discriminative Feature Learning Approach for Deep Face Recognition[C]//Lecture Notes in Computer Science: 9911. Heidelberg: Springer Verlag, 2016: 499-515.

方法	Bicubic	SRCNN^[1]	DRRN^[5]	Wavelet-SRNet^[6]	文中方法
PSNR/dB	21.02	21.63	22.83	24.12	22.40
SSIM	0.577 1	0.624 8	0.706 9	0.778 8	0.704 0

方法	Bicubic	SRCNN^[1]	DRRN^[5]	Wavelet-SRNet^[6]	文中方法
PSNR/dB	25.25	26.12	27.94	29.23	26.26
SSIM	0.783 5	0.825 1	0.879 2	0.909 5	0.863 9

方法	Bicubic	SRCNN^[1]	DRRN^[5]	Wavelet-SRNet^[6]	文中方法
8×8(放大8倍)	64.38	62.81	79.40	87.65	90.18
16×16(放大4倍)	90.21	91.31	93.86	96.15	96.26

结合小波与递归神经网络的低分辨率人脸识别

Low resolution face recognition method based on wavelet and recursive neural networks

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 17

相关文章 15

Metrics

本文评价

推荐阅读 10

[1]	张宇浩,程培涛,张书豪,王秀美. 一种自适应权重学习的轻量超分辨率重建网络[J]. 西安电子科技大学学报, 2021, 48(5): 15-22.
[2]	刘树东,王晓敏,张艳. 一种对称残差CNN的图像超分辨率重建方法[J]. 西安电子科技大学学报, 2019, 46(5): 15-23.
[3]	蒋行国,万今朝,蔡晓东,李海鸥,曹艺. 奇异值分解与中心度量的细粒度车型识别算法[J]. 西安电子科技大学学报, 2019, 46(3): 82-88.
[4]	陈晓范,申海杰,边倩,王振铎,田新志. 结合注意力机制的人脸超分辨率重建[J]. 西安电子科技大学学报, 2019, 46(3): 148-153.
[5]	时文俊,郭从洲,童晓冲,田园,曹闻. 辐射保真的红外遥感图像超分辨率重建[J]. 西安电子科技大学学报, 2019, 46(2): 107-113.
[6]	梁晓曦, 蔡晓东, 王萌, 库浩华. NRCNN与角度度量融合的人脸识别方法[J]. 西安电子科技大学学报, 2018, 45(6): 144-149.
[7]	曾燕;陈岳林;蔡晓东. 结合全局与局部池化的深度哈希人脸识别算法[J]. 西安电子科技大学学报, 2018, 45(5): 163-169.
[8]	魏冬梅;周卫东. 近邻样本协作表示的人脸识别算法[J]. J4, 2015, 42(3): 115-121.
[9]	李云飞;卢朝阳;李静;姚超. 鼻子区域生物特征识别[J]. J4, 2014, 41(4): 20-25+165.
[10]	赵扬扬;周水生;武亚静. 一种用于人脸识别的非迭代GLRAM算法[J]. J4, 2014, 41(2): 144-150.
[11]	侯俊;郝秀娟;谢德燕;高全学. 二维多样性保持投影及人脸识别[J]. J4, 2012, 39(6): 34-41.
[12]	鲁珂;丁正明;赵继东;吴跃. 一种基于相关反馈的视频人脸算法[J]. J4, 2012, 39(3): 154-160.
[13]	田玉敏；云艳娥；马天骏. 判别近邻保持嵌入人脸识别[J]. J4, 2011, 38(3): 24-28+98.
[14]	靳薇;张建奇;梁建霞;王加强. 一种新的人脸归一化算法[J]. J4, 2011, 38(2): 61-65+111.
[15]	温浩;孙蕾. 基于张量局部和全局信息的人脸识别算法[J]. J4, 2010, 37(3): 429-435.