非受限条件下的深度人脸年龄分类

doi:10.11772/j.issn.1001-9081.2017.11.3244

计算机应用 ›› 2017, Vol. 37 ›› Issue (11): 3244-3248.DOI: 10.11772/j.issn.1001-9081.2017.11.3244

• 2017年中国计算机学会人工智能会议(CCFAI 2017) • 上一篇下一篇

非受限条件下的深度人脸年龄分类

张珂, 高策, 郭丽茹, 苑津莎, 赵振兵

华北电力大学电子与通信工程系, 河北保定 071000

收稿日期:2017-05-11 修回日期:2017-05-27 出版日期:2017-11-10 发布日期:2017-11-11
通讯作者: 张珂
作者简介:张珂(1980-),男,河北保定人,副教授,博士,CCF会员,主要研究方向:计算机视觉、深度学习、机器学习、机器人导航、自然语言处理、空间关系描述;高策(1992-),女,河北保定人,硕士研究生,主要研究方向:计算机视觉、深度学习;郭丽茹(1992-),女,河北石家庄人,硕士研究生,主要研究方向:计算机视觉、深度学习;苑津莎(1957-),男,河北保定人,教授,博士,CCF会员,主要研究方向:智能电网信息处理、无线通信、电磁场数值计算;赵振兵(1979-),男,江苏宿迁人,副教授,博士,CCF会员,主要研究方向:图像处理、电力设备智能检测。
基金资助:
国家自然科学基金资助项目（61302163，61302105，61401154）；河北省自然科学基金资助项目（F2015502062，F2016502101）；中央高校基本科研经费资助项目（2016MS99，2015ZD20）。

Deep face age classification under unconstrained conditions

ZHANG Ke, GAO Ce, GUO Liru, YUAN Jinsha, ZHAO Zhenbing

Department of Electronic and Communication Engineering, North China Electric Power University, Baoding Hebei 071000, China

Received:2017-05-11 Revised:2017-05-27 Online:2017-11-10 Published:2017-11-11
Supported by:
This work is partially supported by the National Natural Science Foundation of China (61302163, 61302105, 61401154), the Natural Science Foundation of Hebei Province (F2015502062, F2016502101), and the Fundamental Research Funds for the Central Universities (2016MS99, 2015ZD20).

摘要/Abstract

摘要： 针对非受限条件下人脸图像年龄分类准确度较低的问题，提出了一种基于深度残差网络（ResNets）和大数据集微调的非受限条件下人脸年龄分类方法。首先，选用深度残差网络作为基础卷积神经网络模型处理人脸年龄分类问题；其次，在ImageNet数据集上对深度残差网络预训练，学习基本图像特征的表达；然后，对大规模人脸年龄图像数据集IMDB-WIKI清洗，并建立了IMDB-WIKI-8数据集用于微调深度残差网络，实现一般物体图像到人脸年龄图像的迁移学习，使模型适应于年龄段的分布并提高网络学习能力；最后，在非受限人脸数据集Adience上对微调后的网络模型进行训练和测试，并采用交叉验证方法获取年龄分类准确度。通过34/50/101/152层残差网络对比可知，随着网络层数越深年龄分类准确度越高，并利用152层残差网络获得了Adience数据集上人脸图像年龄分类的最高准确度65.01%。实验结果表明，结合更深层残差网络和大数据集微调，能有效提高人脸图像年龄分类准确度。

关键词: 非受限人脸年龄分类, 深度残差网络, 迁移学习, ImageNet

Abstract: Concerning low accuracy of age classification of face images under unrestricted conditions, a new method of face age classification under unconstrained conditions based on deep Residual Networks (ResNets) and large dataset pre-training was proposed. Firstly, the deep residual networks were used as the basis convolutional neural network model to deal with the problem of face age classification. Secondly, the deep residual networks were trained on the ImageNet dataset to learn the expression of basic image features. Thirdly, the large-scale face age images IMDB-WIKI was cleaned, and the IMDB-WIKI-8 dataset was established for fine-tuning the deep residual networks, and migration learning from the general object image to face age image was achieved to make the model adapt to the distribution of the age group and improve the network learning capability. Finally, the fine-tuned network model was trained and tested on the unconstrained Adience dataset, and the age classification accuracy was obtained by the cross-validation method. Through the comparison of 34/50/101/152-layer residual networks, it could be seen that the more layers of the network have the higher accuracy of age classification. And the best state-of-the-art age classification result on Adience dataset with the accuracy of 65.01% was achieved by using the 152-layer residual network. The experimental results show that the combination of deeper residual network and large dataset pretraining can effectively improve the accuracy of face age classification.

Key words: unconstrained face age classification, deep Residual Networks (ResNets), migrate learning, ImageNet

中图分类号:

TP391.41

张珂, 高策, 郭丽茹, 苑津莎, 赵振兵. 非受限条件下的深度人脸年龄分类[J]. 计算机应用, 2017, 37(11): 3244-3248.

ZHANG Ke, GAO Ce, GUO Liru, YUAN Jinsha, ZHAO Zhenbing. Deep face age classification under unconstrained conditions[J]. Journal of Computer Applications, 2017, 37(11): 3244-3248.

参考文献

[1] 王先梅, 梁玲燕, 王志良, 等. 人脸图像的年龄估计技术研究[J]. 中国图象图形学报, 2012, 17(6):603-618.(WANG X M, LIANG L Y, WANG Z L, et al. Age estimation by facial image:a survey[J]. Journal of Image and Graphics, 2012, 17(6):603-618.)
[2] EIDINGER E, ENBAR R, HASSNER T. Age and gender estimation of unfiltered faces[J]. IEEE Transactions on Information Forensics & Security, 2014, 9(12):2170-2179.
[3] PANIS G, LANITIS A, TSAPATSOULIS N, et al. Overview of research on facial ageing using the FG-NET ageing database[J]. IET Biometrics, 2016, 5(2):37-46.
[4] CHEN B C, CHEN C S, HSU W H. Face recognition and retrieval using cross-age reference coding with cross-age celebrity dataset[J]. IEEE Transactions on Multimedia, 2015, 17(6):804-815.
[5] GUO G. Human Age Estimation and Sex Classification[M]. Berlin:Springer, 2012:101-131.
[6] HAN H, OTTO C, JAIN A K. Age estimation from face images:human vs. machine performance[C]//Proceedings of the 2013 International Conference on Biometrics. Piscataway, NJ:IEEE, 2013:1-8.
[7] 王俊艳, 苏光大, 林行刚. 人脸图像年龄估计[J]. 清华大学学报(自然科学版), 2007, 47(4):526-529.(WANG J Y, SU G D, LIN X G. Face age image estimation[J]. Journal of Tsinghua University (Science and Technology), 2007, 47(4):526-529.)
[8] 林时苗, 毛晓蛟, 杨育彬. 基于人脸图像的年龄估计[J]. 计算机科学, 2015, 42(6):32-36.(LIN S M, MAO X J, YANG Y B. Age estimation based on facial image[J]. Computer Science, 2015, 42(6):32-36.)
[9] KWON Y H. Age classification from facial images[C]//Proceedings of the 1994 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Piscataway, NJ:IEEE, 1994:762-767.
[10] HAYASHI J, YASUMOTO M, ITO H, et al. Age and gender estimation based on wrinkle texture and color of facial images[C]//Proceedings of the 16th International Conference on Pattern Recognition. Piscataway, NJ:IEEE, 2002:405-408.
[11] NAKANO M, YASUKATA F, FUKUMI M. Age classification from face images focusing on edge information[C]//Proceedings of the 8th International Conference on Knowledge-Based and Intelligent Information and Engineering Systems. Berlin:Springer, 2004:898-904.
[12] LANITIS A, DRAGANOVA C, CHRISTODOULOU C. Comparing different classifiers for automatic age estimation[J]. IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), 2004, 34(1):621-628.
[13] COOTES T F, EDWARDS G J, TAYLOR C J. Active appearance models[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2001, 23(6):681-685.
[14] LANITIS A. Comparative evaluation of automatic age-progression methodologies[J]. Journal on Advances in Signal Processing, 2008, 2008(1):Article No. 101.
[15] GUO G, MU G, FU Y, et al. A study on automatic age estimation using a large database[C]//Proceedings of the 2009 IEEE 12th International Conference on Computer Vision. Piscataway, NJ:IEEE, 2009:1986-1991.
[16] CHOI S E, LEE Y J, LEE S J, et al. Age estimation using a hierarchical classifier based on global and local facial features[J]. Pattern Recognition, 2011, 44(6):1262-1281.
[17] GUNAY A, NABIYEV V V. Automatic age classification with LBP[C]//Proceedings of the 23rd International Symposium on Computer and Information Sciences. Piscataway, NJ:IEEE, 2008:1-4.
[18] LEVI G, HASSNCER T. Age and gender classification using convolutional neural networks[C]//Proceedings of the 2015 Computer Vision and Pattern Recognition Workshops. Piscataway, NJ:IEEE, 2015:34-42.
[19] HOU L, SAMARAS D, KURC T M, et al. Neural networks with smooth adaptive activation functions for regression[DB/OL].[2016-11-16]. https://arxiv.org/abs/1608.06557.
[20] EKMEKJI A. Convolutional neural networks for age and gender classification[EB/OL].[2016-11-28]. http://cs231n.stanford.edu/reports/2016/pdfs/003_Report.pdf.
[21] ROTHE R, TIMOFTE R, VAN GOOL L. Deep expectation of real and apparent age from a single image without facial landmarks[EB/OL].[2017-05-10].http://homes.esat.kuleuven.be/~konijn/publications/2016/DEEP-EXP.pdf.
[22] HOU L, YU C P, SAMARAS D. Squared earth mover's distance-based loss for training deep neural networks[EB/OL].[2017-05-10].https://arxiv.org/pdf/1611.05916.pdf.
[23] DEHGHAN A, ORTIZ E G, SHU G, et al. DAGER:Deep age, gender and emotion recognition using convolutional neural network[EB/OL].[2017-03-16].https://arxiv.org/pdf/1702.04280.pdf.
[24] LECUN Y, BOTTOU L, BENGIO Y, et al. Gradient-based learning applied to document recognition[J]. Proceedings of the IEEE, 1998, 86(11):2278-2324.
[25] KRIZHEVSKY A, SUTSKEVER I, HINTON G E. ImageNet classification with deep convolutional neural networks[C]//Proceedings of the 25th International Conference on Neural Information Processing Systems. Lake Tahoe, Nevada:Curran Associates Inc., 2012:1097-1105.
[26] SIMONYAN K, ZISSERMAN A. Very deep convolutional networks for large-scale image recognition[EB/OL].[2016-12-21]. https://arxiv.org/abs/1409.1556.
[27] SZEGEDY C, LIU W, JIA Y, et al. Going deeper with convolutions[C]//Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ:IEEE, 2015:1-9.
[28] HE K, ZHANG X, REN S, et al. Deep residual learning for image recognition[C]//Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ:IEEE, 2016:770-778.
[29] CUBBEE, GROSS S. ResNet models trained on ImageNet[EB/OL].[2016-12-10]. https://github.com/facebook/fb.resnet.torch/tree/master/pretrained.

非受限条件下的深度人脸年龄分类

Deep face age classification under unconstrained conditions

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	陈争涛, 黄灿, 杨波, 赵立, 廖勇. 基于迁移学习的并行卷积神经网络牦牛脸识别算法[J]. 计算机应用, 2021, 41(5): 1332-1336.
[2]	任奕茗, 王让定, 严迪群, 林昱臻. 基于深度残差网络的语音隐写分析方法[J]. 计算机应用, 2021, 41(3): 774-779.
[3]	王锦凯, 贾旭. 基于迁移孪生非负矩阵分解的静脉识别算法[J]. 计算机应用, 2021, 41(3): 898-903.
[4]	钟莎, 黄玉清. 基于孪生区域候选网络的无人机指定目标跟踪[J]. 计算机应用, 2021, 41(2): 523-529.
[5]	陈朗, 王让定, 严迪群, 林昱臻. 融合残差网络和极限梯度提升的音频隐写检测模型[J]. 计算机应用, 2021, 41(2): 449-455.
[6]	曹建芳, 闫敏敏, 贾一鸣, 田晓东. 融合迁移学习的Inception-v3模型在古壁画朝代识别中的应用[J]. 《计算机应用》唯一官方网站, 2021, 41(11): 3219-3227.
[7]	赵津, 宋文爱, 邰隽, 杨吉江, 王青, 李晓丹, 雷毅, 邱悦. 儿童阻塞性睡眠呼吸暂停计算机人脸辅助诊断综述[J]. 《计算机应用》唯一官方网站, 2021, 41(11): 3394-3401.
[8]	刘晓龙, 王士同. 渐进式分离的开放集模糊域自适应算法[J]. 《计算机应用》唯一官方网站, 2021, 41(11): 3127-3131.
[9]	黄学雨, 徐浩特, 陶剑文. 具有特征选择的多源自适应分类框架[J]. 计算机应用, 2020, 40(9): 2499-2506.
[10]	张晨曦, 唐曙, 唐珂. 迁移学习下的火箭发动机参数异常检测策略[J]. 计算机应用, 2020, 40(9): 2774-2780.
[11]	白茹意, 郭小英, 贾春花. 基于特征融合的小样本抽象画图像情感预测[J]. 计算机应用, 2020, 40(8): 2207-2213.
[12]	马金林, 魏萌, 马自萍. 基于深度迁移学习的肺结节分割方法[J]. 计算机应用, 2020, 40(7): 2117-2125.
[13]	肖禾, 刘志勤, 王庆凤, 黄俊, 周莹, 刘启榆, 徐卫云. 基于迁移学习的多视角乳腺肿块和钙化簇分类方法[J]. 计算机应用, 2020, 40(5): 1460-1464.
[14]	陈杰, 张挺, 杜奕. 基于自适应深度迁移学习的多孔介质重构[J]. 计算机应用, 2020, 40(4): 1231-1236.
[15]	谌贵辉, 易欣, 李忠兵, 钱济人, 陈伍. 基于改进YOLOv2和迁移学习的管道巡检航拍图像第三方施工目标检测[J]. 计算机应用, 2020, 40(4): 1062-1068.