基于径向基网络的无监督亲子关系验证

张鑫坤; 范永胜

doi:10.13568/j.cnki.651094.651316.2023.02.22.0001

您当前的位置：

首页 >

文章列表页 >

基于径向基网络的无监督亲子关系验证

更新时间：2026-01-21

- 基于径向基网络的无监督亲子关系验证
- Journal of Xinjiang University (Natural Science Edition in Chinese and English) Vol. 40, Issue 4, Pages: 453-460(2023)
- 作者机构：
  
  重庆师范大学计算机与信息科学学院
- 作者简介：
- 基金信息：
- DOI：10.13568/j.cnki.651094.651316.2023.02.22.0001
  CLC： TP391.41
- Published：2023
- 稿件说明：
移动端阅览
[1]张鑫坤,范永胜.基于径向基网络的无监督亲子关系验证[J].新疆大学学报(自然科学版)(中英文),2023,40(04):453-460.
[1]张鑫坤,范永胜.基于径向基网络的无监督亲子关系验证[J].新疆大学学报(自然科学版)(中英文),2023,40(04):453-460. DOI： 10.13568/j.cnki.651094.651316.2023.02.22.0001.

DOI：10.13568/j.cnki.651094.651316.2023.02.22.0001.

摘要

亲子关系验证旨在通过给定的两张图像判断其是否具有亲子关系，该问题在机器视觉领域日益受到人们的关注．现有的方法将工作重心放在学习更有效的度量、提取更有判别力的特征、选择更好的分类器上，无法较好地解决成对图像之间差异大的问题．提出了基于径向基网络（RBFN）的无监督分类方法，该方法首先将图像进行聚类，然后将聚类中心点作为径向基核函数的中心点，利用RBFN在成对图像之间建立多变量插值函数，为了使函数逼近的效果更好，将RBFN的输出与子女方图像计算损失并将损失反向传播，不断优化网络模型，最后根据阈值完成分类．算法在公共数据集KinFaceW和UB KinFace上进行了5折交叉验证，结果充分证明了所提方法的有效性及对样本间差异的适应性，为后续亲子关系验证研究提供参考．

Abstract

Kinship verification aims to verify whether kinship exists in the given two images. The previous methods focus on feature extraction or selecting better classifiers

yet they can not handle the barrier of huge difference between paired images. This paper proposed an unsupervised classification method based on radial basis function network(RBFN)

which firstly clustered the image

and then took the cluster center points as the center points of the radial basis function

then used the RBFN to establish a multivariable interpolation function between the pair images. In order to make the function approximation better

the output of RBFN is used to calculate the MSE loss with the children's images

and the loss is back-propagated to continuously optimize the network model

finally completed the classification according to the threshold. 5-Fold cross validation experiments on two publicly available kinship datasets Kin FaceW and UB KinFace demonstrate the effectiveness of our proposed method and its adaptability to the differences.

关键词

Keywords

references

DAL M M F, MALONEY L T. Lateralization of kin recognition signals in the human face[J]. Journal of Vision, 2010, 10(8):9.

FANG R G, TANG K D, SNAVELY N, et al. Towards computational models of kinship verification[C]//Proceedings of the International Conference on Image Processing. Piscataway:IEEE, 2010.

LU J W, ZHOU X Z, TAN Y P, et al. Neighborhood repulsed metric learning for kinship verification[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2014, 36(2):331-345.

ZHOU X Z, JIN K, XU M, et al. Learning deep compact similarity metric for kinship verification from face images[J]. Information Fusion, 2018, 48:84-94.

HU J L, LU J W, TAN Y P, et al. Local large-margin multi-metric learning for face and kinship verification[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2017, 28(8):1875-1891.

LIU H J, ZHU C. Status-aware projection metric learning for kinship verification[C]//IEEE International Conference on Multimedia and Expo. Piscataway:IEEE, 2017.

WEI Z Q, XU M, GENG L, et al. Adversarial similarity metric learning for kinship verification[J]. IEEE Access, 2019, 7:100029-100035.

ZHAO Y G, SONG Z, ZHENG F, et al. Learning a multiple kernel similarity metric for kinship verification[J]. Information Sciences,2018, 430/431:247-260.

ZHOU X Z, SHANG Y Y, YAN H B, et al. Ensemble similarity learning for kinship verification from facial images in the wild[J].Information Fusion, 2016, 32:40-48.

QIAO Y X, CUI J B, HUANG F X, at al. Efficient style-corpus constrained learning for photorealistic style transfer[J]. IEEE Transactions on Image Processing, 2021, 30:3154-3166.

FAN B, LU H M, ZENG H, et al. Deep unsupervised binary descriptor learning through locality consistency and self distinctiveness[J]. IEEE Transactions on Multimedia, 2021, 23:2770-2781.

孟欣欣,阿里甫·库尔班,吕情深,等.基于迁移学习的自然环境下香梨目标识别研究[J].新疆大学学报(自然科学版), 2019, 36(4):461-467.

LU J W, HU J L, TAN Y P. Discriminative deep metric learning for face and kinship verification[J]. IEEE Transactions on Image Processing, 2017, 26(9):4269-4282.

CHERGUI A, OUCHTATI S, MAVROMATIS S, et al. Kinship verification through facial images using CNN-based features[J].Traitement du Signal, 2020, 37(1):1-8.

郑亮,陈亚,陈小潘,等.人脸局部特征增强的亲子关系验证方法[J].计算机工程与应用, 2021, 57(23):226-233.

YAN H B, SONG C H. Multi-scale deep relational reasoning for facial kinship verification[J]. Pattern Recognition, 2020, 110(2):107541.

NANDY A, MONDAL S S. Kinship verification using deep siamese convolutional neural network[C]//2019 14th IEEE International Conference on Automatic Face&Gesture Recognition(FG 2019). Lille:IEEE, 2019.

LI W H, LU J W, FENG J J, et al. Reasoning graph networks for kinship verification:from star-shaped to hierarchical[J]. IEEE Transactions on Image Processing, 2021, 30:4947-4961.

SERRAOUI I, LAIADI O, OUAMANE A, et al. Knowledge-based tensor subspace analysis system for kinship verification[J].Neural Networks, 2022, 151:222-237.

AMINA T, ABDELMALIK T, DJAMEL S, et al. Deep learning features for robust facial kinship verification[J]. IET Image Processing, 2018, 12(12):2336-2345.

ZHANG L, DUAN Q Y, ZHANG D, et al. AdvKin:adversarial convolutional network for kinship verification[J]. IEEE Transactions on Cybernetics, 2021, 51(12):5883-5896.

SELLAM A, AZZOUNE H. Neighborhood min distance descriptor for kinship verification[J]. Multimedia Tools and Applications,2020, 79(12):1-20.

LU J W, HU J L, ZHOU X Z, et al. Neighborhood repulsed metric learning for kinship verification[C]//Computer Vision and Pattern Recognition. Indonesia:IEEE, 2012.

XIA S Y, SHAO M, LUO J B, et al. Understanding kin relationships in a photo[J]. IEEE Transactions on Multimedia, 2012,14(4):1046-1056.

BEKHOUCHE S, OUAFI A, DORNAIKA F, et al. Pyramid multi-level features for facial demographic estimation[J]. Expert Systems with Applications, 2017, 80:297-310.

PAUL V, MICHAEL J. Robust real-time face detection[J]. International Journal of Computer Vision, 2004, 57(2):137-154.

ZHANG K P, ZHANG Z P, LI Z F, et al. Joint face detection and alignment using multitask cascaded convolutional networks[J].IEEE Signal Processing Letters, 2016, 23(10):1499-1503.

KOHLI N, VATSA M, SINGH R, et al. Hierarchical representation learning for kinship verification[J]. IEEE Transactions on Image Processing, 2017, 26(1):289-302.

YAN H B, LU J W, DENG W H, et al. Discriminative multimetric learning for kinship verification[J]. IEEE Transactions on Information Forensics and Security, 2014, 9(7):1169-1178.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

新疆棉花主栽品种的分类与优势比较

新疆小蜂总科资源及其在生物防治中的应用前景

Related Author

孙起

刘瑞

王立永

肖远淑

卫军

贾丽霞

康宁

胡红英

Related Institution

新疆大学纺织与服装学院

东华大学纺织学院

新疆维吾尔自治区纤维检验局

新疆大学生命科学与技术学院新疆生物资源基因工程重点实验室

⁰