基于标记分布学习的人脸年龄识别算法

张会影; 圣文顺; 曾耀徵

doi:10.3969/j.issn.1671-7775.2023.02.008

PDF(2281 KB)

江苏大学学报（自然科学版） ›› 2023, Vol. 44 ›› Issue (2) : 180-185. DOI: 10.3969/j.issn.1671-7775.2023.02.008

论文

基于标记分布学习的人脸年龄识别算法

作者信息 +

Face age recognition algorithm based on label distribution learning

Author information +

文章历史 +

摘要

针对人脸年龄识别可用数据集普遍不足的问题，为提升可用数据集不变情况下人脸年龄识别的精度，在深度学习（DL）框架中引入标记分布学习（LDL）策略，命名为DL-LDL，其中卷积神经网络用于自动提取人脸特征，改进的标记分布学习用于学习真实年龄及相邻年龄之间的模糊性和多义性，以丰富年龄信息，提高识别精度.将DL-LDL方法在MORPH和FG-NET这2个公开数据集上进行了试验测试.结果表明：DL-LDL方法提高了年龄识别的精度，与现有最先进的人脸年龄识别方法相比，在MORPH和FG-NET上的平均绝对误差分别降低了8.2%和13.8%.

Abstract

To solve the problem that the available datasets for face age recognition were generally insufficient, and improve the accuracy of face age recognition with the same available dataset, the label distribution learning (LDL) strategy was introduced into the deep learning (DL) framework and named as DL-LDL. Convolutional neural networks were used to automatically extract facial features, and the improved label distribution learning was used to learn the fuzziness and ambiguity between the real age and the adjacent ages for enriching the age information and improving the recognition accuracy. The DL-LDL method was tested on two open datasets of MORPH and FG-NET. The results show that the DL-LDL method can improve the accuracy of age recognition. Compared with the most advanced facial age recognition method at present, the mean absolute errors of MORPH and FG-NET are reduced by 8.2% and 13.8%, respectively.

导出引用

张会影, 圣文顺, 曾耀徵. 基于标记分布学习的人脸年龄识别算法[J]. 江苏大学学报（自然科学版）, 2023, 44(2): 180-185 https://doi.org/10.3969/j.issn.1671-7775.2023.02.008

ZHANG Huiying, SHENG Wenshun, ZENG Yaozheng. Face age recognition algorithm based on label distribution learning[J]. Journal of Jiangsu University(Natural Science Edition), 2023, 44(2): 180-185 https://doi.org/10.3969/j.issn.1671-7775.2023.02.008

参考文献

[1]ZHANG B C, BAO Y. Cross-dataset learning for age estimation[J]. IEEE Access, 2022(10): 24048-24055.

[2]齐妙,闫光友,徐慧,等.基于多尺度特征选择网络的人脸表情识别[J].吉林大学学报(理学版),2022,60(2):425-431.

QI M, YAN G Y, XU H, et al. Facial expression re-cognition based on multi-scale feature selection network [J]. Journal of Jilin University (Science Edition),2022,60(2):425-431.（in Chinese）

[3]李军,李明.改进多尺度卷积神经网络的人脸表情识别研究[J].重庆邮电大学学报(自然科学版),2022,34(2):201-207.

LI J, LI M.Research on facial expression recognition based on improved multi-scale convolutional neural networks [J]. Journal of Chongqing University of Posts and Telecommunications (Natural Science Edition),2022, 34(2): 201-207.（in Chinese）

[4]王新月,钟福金.基于多分支CNN和多尺度特征融合的非受控人脸年龄估计[J].重庆邮电大学学报(自然科学版),2022,34(4):612-620.

WANG X Y, ZHONG F J. Uncontrolled facial age estimation based on multi-branch CNN and multi-scale feature fusion [J]. Journal of Chongqing University of Posts and Telecommunications (Natural Science Edition),2022,34(4): 612-620.（in Chinese）

[5]GAO B B, XING C, XIE C W, et al. Deep label distribution learning with label ambiguity[J]. IEEE Transactions on Image Processing, 2017,26(6):2825-2838.

[6]GENG X, SMITH-MILES K, ZHOU Z H. Facial age estimation by learning from label distributions[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013,35(10):2401-2412.

[7]GENG X, WANG Q, XIA Y. Facial age estimation by adaptive label distribution learning[C]∥Proceedings of the 2014 22nd International Conference on Pattern Re-cognition, 2014:4465-4470.

[8]NG C C, YAP M H, CHENG Y T, et al. Hybrid ageing patterns for face age estimation [J]. Image and Vision Computing, 2018,69: 92-102.

[9]ROTHE R, TIMOFTE R, VAN GOOL L. Deep expectation of real and apparent age from a single image without facial landmarks[J]. International Journal of Computer Vision,2018,126(2/3/4):144-157.

[10]RIZWAN S A, GHADI Y Y, JALAL A, et al. Automated facial expression recognition and age estimation using deep learning[J]. Computers, Materials and Continua, 2022,71(2):5235-5252.

[11]DAGHER I, BARBARA D. Facial age estimation using pre-trained CNN and transfer learning[J]. Multimedia Tools and Applications,2021,80(13):20369-20380.