MD-FVR： cascaded finger vein recognition network based on multi-domain feature fusion

doi:10.11772/j.issn.1001-9081.2025050658

Journal of Computer Applications ›› 2026, Vol. 46 ›› Issue (5): 1658-1666.DOI: 10.11772/j.issn.1001-9081.2025050658

• Frontier and comprehensive applications • Previous Articles

MD-FVR： cascaded finger vein recognition network based on multi-domain feature fusion

Chi ZHANG¹, Xianjing MENG¹(), Changhao DOU¹, Qian WANG¹, Leilei GENG¹, Xiaoming XI²

^1.School of Computing and Artificial Intelligence，Shandong University of Finance and Economics，Jinan Shandong 250014，China
^2.School of Computer Science and Technology，Shandong Jianzhu University，Jinan Shandong 250101，China

Received:2025-06-16 Revised:2025-08-01 Accepted:2025-08-08 Online:2025-08-15 Published:2026-05-10
Contact: Xianjing MENG
About author:ZHANG Chi， born in 2001， M. S. candidate. His research interests include finger vein recognition.
DOU Changhao， born in 2000， M. S. candidate. His research interests include compositional zero-shot learning.
WANG Qian， born in 1976， Ph. D. associate professor. Her research interests include machine learning， intelligent communication and networking.
GENG Leilei， born in 1984， Ph. D.， associate professor. Her research interests include digital image processing.
XI Xiaoming， born in 1987， Ph. D.， professor. His research interests include medical image processing， machine learning.
Supported by:
General Program of Natural Science Foundation of Shandong Province(ZR2023MF075)

MD-FVR：基于多域特征融合的级联手指静脉识别网络

张驰¹, 孟宪静¹(), 窦长昊¹, 王倩¹, 耿蕾蕾¹, 袭肖明²

^1.山东财经大学计算机与人工智能学院，济南 250014
^2.山东建筑大学计算机与人工智能学院，济南 250101

通讯作者: 孟宪静
作者简介:张驰（2001—），男，山东菏泽人，硕士研究生，CCF会员，主要研究方向：手指静脉识别
窦长昊（2000—），男，山东聊城人，硕士研究生，CCF会员，主要研究方向：组合零样本学习
王倩（1976—），女，山东济南人，副教授，博士，CCF会员，主要研究方向：机器学习、智能通信与网络
耿蕾蕾（1984—），女，山东滕州人，副教授，博士，CCF会员，主要研究方向：数字图像处理
袭肖明（1987—），男，山东济南人，教授，博士，CCF会员，主要研究方向：医学图像处理、机器学习。
基金资助:
山东省自然科学基金面上项目(ZR2023MF075);山东省自然科学基金面上项目(ZR2023MF039)

Abstract

Abstract:

Existing finger vein recognition methods primarily focus on extracting features from the spatial domain， while neglecting the frequency-domain representation of tubular tree-like veins and the embedded multi-scale details. To address this issue， a cascaded Finger Vein Recognition network based on Multi-Domain feature fusion （MD-FVR） was proposed. The method adopted a hierarchical cascaded network architecture， in which each stage consisted of a multi-domain feature fusion module and basic operations， enabling progressive feature extraction and enhancement. The core multi-domain feature fusion module operated as follows： firstly， structural information was enhanced through Depthwise Separable Wavelet Convolution blocks （DSWC）. Secondly， frequency-domain global features were extracted using Frequency-Domain Spatial and Coupling blocks （FDSCs） to compensate for limitations in spatial domain representation. Combined with an improved Half-Wavelet Spatial Attention block （HWSA）， the local frequency-domain details were further refined. Finally， the final enhanced feature representation was obtained by integrating multi-domain features through the Adaptive Feature Fusion module （AFF）. Experimental results on the HKPU and SDUMLA-HMT datasets showed that MD-FVR achieved recognition accuracies of 99.68% and 99.53%， with Equal Error Rates （EERs） of 0.35% and 0.41%， respectively. Compared with methods that rely solely on spatial or frequency features， MD-FVR demonstrates significant improvements in both recognition accuracy and robustness.

Key words: finger vein recognition, deep neural network, multi-domain feature fusion, frequency domain analysis, spatial-wavelet attention, cascaded network architecture

摘要：

针对现有手指静脉识别方法主要关注空间域特征的提取，忽略了手指静脉血管作为管状树形结构在频率域的表达及其中蕴含的多尺度细节的问题，提出基于多域特征融合的级联手指静脉识别网络（MD-FVR），采用层次化的级联网络架构，每一层次由多域融合模块与基础操作构成，能实现特征的逐步提取和增强。其中，多域特征融合模块作为核心模块：首先，通过深度可分离小波卷积块（DSWC）增强结构信息；其次，利用频率域-空间耦合块（FDSC）提取频率域全局特征，弥补空间域表达的不足，并结合改进的半小波-空间注意力块（HWSA）进一步细化频率域细节的表征；最后，通过自适应特征融合模块（AFF）整合多域特征，得到最终的增强特征表示。在HKPU与SDUMLA-HMT数据集上的实验结果显示，MD-FVR识别准确率分别达到了99.68%和99.53%，且等错误率（EER）仅为0.35%和0.41%。与现有的仅在空间域或频率域提取特征的方法相比，MD-FVR方法在识别准确性与鲁棒性方面均表现出显著的优势。

关键词: 指静脉识别, 深度神经网络, 多域特征融合, 频率域分析, 空间域-小波注意力, 级联网络架构

CLC Number:

TP391.41

Chi ZHANG, Xianjing MENG, Changhao DOU, Qian WANG, Leilei GENG, Xiaoming XI. MD-FVR： cascaded finger vein recognition network based on multi-domain feature fusion[J]. Journal of Computer Applications, 2026, 46(5): 1658-1666.

张驰, 孟宪静, 窦长昊, 王倩, 耿蕾蕾, 袭肖明. MD-FVR：基于多域特征融合的级联手指静脉识别网络[J]. 《计算机应用》唯一官方网站, 2026, 46(5): 1658-1666.

Figures/Tables 12

References 43

[1]	KOLIVAND H， ASADIANFAM S， AKINTOYE K A， et al. Finger vein recognition techniques： a comprehensive review［J］. Multimedia Tools and Applications， 2023， 82（22）： 33541-33575.
[2]	ROSDI B A， SHING C W， SUANDI S A. Finger vein recognition using local line binary pattern［J］. Sensors， 2011， 11（12）： 11357-11371.
[3]	MENG X， YANG G， YIN Y， et al. Finger vein recognition based on local directional code［J］. Sensors， 2012， 12（11）： 14937-14952.
[4]	YANG J， SHI Y. Finger-vein ROI localization and vein ridge enhancement［J］. Pattern Recognition Letters， 2012， 33（12）： 1569-1579.
[5]	XI X， YANG L， YIN Y. Learning discriminative binary codes for finger vein recognition［J］. Pattern Recognition， 2017， 66： 26-33.
[6]	ZHANG G， MENG X. High security finger vein recognition based on robust keypoint correspondence clustering［J］. IEEE Access， 2021， 9： 154058-154070.
[7]	YU C B， QIN H F， ZHANG L， et al. Finger-vein image recognition combining modified Hausdorff distance with minutiae feature matching［J］. Journal of Biomedical Science and Engineering， 2009， 2： 261-272.
[8]	KIM H， LEE E J， YOON G J， et al. Illumination normalization for SIFT based finger vein authentication［C］// Proceedings of the 2012 International Symposium on Visual Computing， LNCS 7432. Berlin： Springer， 2012： 21-30.
[9]	SONG W， KIM T， KIM H C， et al. A finger-vein verification system using mean curvature［J］. Pattern Recognition Letters， 2011， 32（11）： 1541-1547.
[10]	CHO S R， PARK Y H， NAM G P， et al. Enhancement of finger-vein image by vein line tracking and adaptive Gabor filtering for finger-vein recognition［J］. Applied Mechanics and Materials， 2012， 145： 219-223.
[11]	DAMAVANDINEJADMONFARED S， MOBARAKEH A K， PASHNA M， et al. Finger vein recognition using PCA-based methods［J］. International Scholarly and Scientific Research and Innovation， 2012， 6（6）： 593-595.
[12]	YANG G， XI X， YIN Y. Finger vein recognition based on （2D）2 PCA and metric learning［J］. BioMed Research International， 2012， 2012： No.324249.
[13]	WU J D， LIU C T. Finger-vein pattern identification using SVM and neural network technique［J］. Expert Systems with Applications， 2011， 38（11）： 14284-14289.
[14]	PARK K R. Finger vein recognition by combining global and local features based on SVM［J］. Computing and Informatics， 2012， 30（2）： 295-309.
[15]	ZHANG Z B， WU D Y， MA S L， et al. Multiscale feature extraction of finger-vein patterns based on wavelet and local interconnection structure neural network［C］// Proceedings of the 2005 International Conference on Neural Networks and Brain. Piscataway： IEEE， 2005： 1081-1084.
[16]	LU Y， XIE S J， YOON S， et al. Robust finger vein ROI localization based on flexible segmentation［J］. Sensors， 2013， 13（11）： 14339-14366.
[17]	HUANG Y， MA H， WANG M. Axially-enhanced local attention network for finger vein recognition［J］. IEEE Transactions on Instrumentation and Measurement， 2023， 72： No.5020210.
[18]	QI Y， HE Y， QI X， et al. Dynamic snake convolution based on topological geometric constraints for tubular structure segmentation［C］// Proceedings of the 2023 IEEE/CVF International Conference on Computer Vision. Piscataway： IEEE， 2023： 6047-6056.
[19]	PAN Z， WANG J， WANG G， et al. Multi-scale deep representation aggregation for vein recognition［J］. IEEE Transactions on Information Forensics and Security， 2021， 16： 1-15.
[20]	KUZU R S， PICIUCCO E， MAIORANA E， et al. On-the-fly finger-vein-based biometric recognition using deep neural networks［J］. IEEE Transactions on Information Forensics and Security， 2020， 15： 2641-2654.
[21]	HUANG J， TU M， YANG W， et al. Joint attention network for finger vein authentication［J］. IEEE Transactions on Instrumentation and Measurement， 2021， 70： No.2513911.
[22]	YANG W， LUO W， KANG W， et al. FVRAS-Net： an embedded Finger-Vein Recognition and AntiSpoofing system using a unified CNN［J］. IEEE Transactions on Instrumentation and Measurement， 2020， 69（11）： 8690-8701.
[23]	SHAHREZA O， MARCEL S. Towards protecting and enhancing vascular biometric recognition methods via biohashing and deep neural networks［J］. IEEE Transactions on Biometrics， Behavior， and Identity Science， 2021， 3（3）： 394-404.
[24]	QIN H， EL-YACOUBI M A. Deep representation for finger-vein image quality assessment［J］. IEEE Transactions on Circuits and Systems for Video Technology， 2018， 28（8）： 1677-1693.
[25]	REN H， SUN L， GUO J， et al. Finger vein recognition system with template protection based on convolutional neural network［J］. Knowledge-Based Systems， 2021， 227： No.107159.
[26]	HOU B， YAN R. ArcVein-arccosine center loss for finger-vein verification［J］. IEEE Transactions on Instrumentation and Measurement， 2021， 70： No.5007411.
[27]	YANG J， YAN M. An improved method for finger-vein image enhancement［C］// Proceedings of the IEEE 10th International Conference on Signal Processing. Piscataway： IEEE， 2010： 1706-1709.
[28]	HUANG J， ZHENG A， SHAKEEL M S， et al. FVFSNet： frequency-spatial coupling network for finger vein authentication［J］. IEEE Transactions on Information Forensics and Security， 2023， 18： 1322-1334.
[29]	KUMAR A， ZHOU Y. Human identification using finger images［J］. IEEE Transactions on Image Processing， 2012， 21（4）： 2228-2244.
[30]	YIN Y， LIU L， SUN X. SDUMLA-HMT： a multimodal biometric database［C］// Chinese Conference on Biometric Recognition， LNCS 7098. Berlin： Springer， 2011： 260-268.
[31]	MENG X J， XI X M， YANG G P， et al. Finger vein recognition based on deformation information［J］. SCIENCE CHINA Information Sciences， 2018， 61（5）： No.052103.
[32]	KAMARUDDINN M， ROSDI B A. A new filter generation method in PCANet for finger vein recognition［J］. IEEE Access， 2019， 7： 132966-132978.
[33]	ZHONG Y， LI J， CHAI T， et al. Different dimension issues in deep feature space for finger-vein recognition［C］// Proceedings of the 2021 Chinese Conference on Biometric Recognition， LNCS 12878. Cham： Springer， 2021： 295-303.
[34]	LI Y， LU H， WANG Y， et al. ViT-Cap： a novel vision transformer-based capsule network model for finger vein recognition［J］. Applied Sciences， 2022， 12（20）： No.10364.
[35]	LI M， GONG Y， ZHENG Z. Finger vein identification based on large kernel convolution and attention mechanism［J］. Sensors， 2024， 24（4）： No.1132.
[36]	TAHIR Y S， ROSDI B A. FV-EffResNet： an efficient lightweight convolutional neural network for finger vein recognition［J］. PeerJ Computer Science， 2024， 10： No.e1837.
[37]	GUO Z， MA H， LI A. A lightweight finger multimodal recognition model based on detail optimization and perceptual compensation embedding［J］. Computer Standards and Interfaces， 2025， 92： No.103937.
[38]	REN H， SUN L， REN J， et al. FV-DDC： a novel finger-vein recognition model with deformation detection and correction［J］. Biomedical Signal Processing and Control， 2025， 100（Pt A）： No.107098.
[39]	SONG J M， KIM W， PARK K R. Finger-vein recognition based on deep DenseNet using composite image［J］. IEEE Access， 2019， 7： 66845-66863.
[40]	ZHANG J， LU Z， LI M， et al. GAN-based image augmentation for finger-vein biometric recognition［J］. IEEE Access， 2019， 7： 183118-183132.
[41]	KUZU R S， MAIORANA E， CAMPISI P. Loss functions for CNN-based biometric vein recognition［C］// Proceedings of the 28th European Signal Processing Conference. Piscataway： IEEE， 2021： 750-754.
[42]	NOH K J， CHOI J， HONG J S， et al. Finger-vein recognition using heterogeneous databases by domain adaption based on a cycle-consistent adversarial network［J］. Sensors， 2021， 21（2）： No.524.
[43]	HOU B， YAN R. Triplet-classifier GAN for finger-vein verification［J］. IEEE Transactions on Instrumentation and Measurement， 2022， 71： No.2505112.

层次	输入尺寸	输出尺寸	输入的通道数	输出的通道数
Stage 1	128×128	64×64	3	16
Stage 2	64×64	32×32	16	64
Stage 3	32×32	16×16	64	256
Stage 4	16×16	8×8	256	1 024

层次	输入尺寸	输出尺寸	输入的通道数	输出的通道数
Stage 1	128×128	64×64	3	16
Stage 2	64×64	32×32	16	64
Stage 3	32×32	16×16	64	256
Stage 4	16×16	8×8	256	1 024

消融模块	HKPU		SDUMLA-HMT
消融模块	ACC	EER	ACC	EER
MD-FVR	99.68	0.35	99.53	0.41
DSWC	99.13	0.56	99.21	0.85
HWSA	98.56	0.54	99.21	0.78
FDSC	99.36	0.75	99.27	0.95
CPCA	98.93	0.43	99.06	0.86

消融模块	HKPU		SDUMLA-HMT
消融模块	ACC	EER	ACC	EER
MD-FVR	99.68	0.35	99.53	0.41
DSWC	99.13	0.56	99.21	0.85
HWSA	98.56	0.54	99.21	0.78
FDSC	99.36	0.75	99.27	0.95
CPCA	98.93	0.43	99.06	0.86

训练样本数	SDUMLA-HMT数据集		HKPU数据集
训练样本数	ACC/%	EER/%	ACC/%	EER/%
3	94.23	3.46	98.71	1.18
4	99.53	0.41	99.52	0.47
5	98.89	1.02	99.68	0.35

MD-FVR： cascaded finger vein recognition network based on multi-domain feature fusion

MD-FVR：基于多域特征融合的级联手指静脉识别网络

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 12

References 43

Related Articles 15

Recommended Articles

Metrics

层次数	HKPU		SDUMLA-HMT
层次数	ACC/%	EER/%	ACC/%	EER/%
3	99.26	0.82	99.15	0.87
4	99.68	0.35	99.53	0.41
5	99.45	0.59	99.32	0.61

网络	SDUMLA-HMT	HKPU
PCANet^［32］	98.19	—
DFCNN^［33］	92.22	96.98
ResNet+SE^［25］	96.70	99.04
ECA-Resnet^［26］	99.25	99.07
ViT-Cap^［34］	93.24	95.61
ALANet^［17］	94.50
LetNet^［35］	99.50	96.10
FV-EffResNet^［36］	98.45	—
LightNet^［37］	95.95	97.33
FV-DDC^［38］	98.74	99.62
MD-FVR	99.53	99.68

网络	SDUMLA-HMT	HKPU
DenseNet-161^［39］	2.35	0.33
FCGAN^［40］	0.87	0.52
Densenet-161^［41］	0.02	1.87
ResNet+SE^［25］	2.14	0.28
ECA-Resnet^［26］	1.53	1.30
CycleGAN+DenseNet161^［42］	3.40	0.85
Triplet-Classifier GAN^［43］	1.33	0.40
ViT-Cap^［34］	1.30	1.66
FVFSNet^［28］	1.10	0.81
LetNet^［35］	0.15	1.21
FV-EffResNet^［36］	0.43	—
LightNet^［37］	0.43	0.51
MD-FVR	0.41	0.35

[1]	Xiaobo QI, Jing ZHANG, Ying SHI, Hui QI, Hangyuan DU. Multiple active learning method based on concept drift detection [J]. Journal of Computer Applications, 2026, 46(5): 1388-1396.
[2]	Rifeng ZHANG, Guangming LI, Yurong OUYANG. Low-light image enhancement network guided by reflection prior map [J]. Journal of Computer Applications, 2026, 46(2): 546-554.
[3]	Ning CAO, Xin WEN, Yanrong HAO, Rui CAO. Lightweight motor imagery electroencephalogram decoding neural network with multi-domain feature fusion [J]. Journal of Computer Applications, 2026, 46(1): 289-296.
[4]	Erkang XIANG, Rong HUANG, Aihua DONG. Open set recognition method with open generation and feature optimization [J]. Journal of Computer Applications, 2025, 45(7): 2195-2202.
[5]	Qiaoling QI, Xiaoxiao WANG, Qianqian ZHANG, Peng WANG, Yongfeng DONG. Label noise adaptive learning algorithm based on meta-learning [J]. Journal of Computer Applications, 2025, 45(7): 2113-2122.
[6]	Huibin WANG, Zhan’ao HU, Jie HU, Yuanwei XU, Bo WEN. Time series forecasting model based on segmented attention mechanism [J]. Journal of Computer Applications, 2025, 45(7): 2262-2268.
[7]	Lu CHEN, Huaiyao WANG, Jingyang LIU, Tao YAN, Bin CHEN. Robotic grasp detection with feature fusion of spatial-Fourier domain information under low-light environments [J]. Journal of Computer Applications, 2025, 45(5): 1686-1693.
[8]	Huahua WANG, Zijian FAN, Ze LIU. Image adversarial example generation method based on multi-space probability enhancement [J]. Journal of Computer Applications, 2025, 45(3): 883-890.
[9]	Handa MA, Yadong WU. Multi-domain spatiotemporal hierarchical graph neural network for air quality prediction [J]. Journal of Computer Applications, 2025, 45(2): 444-452.
[10]	Benchen YANG, Haoran LI, Haibo JIN. Multi-focus image fusion network with cascade fusion and enhanced reconstruction [J]. Journal of Computer Applications, 2025, 45(2): 594-600.
[11]	Sheng YANG, Yan LI. Contrastive knowledge distillation method for object detection [J]. Journal of Computer Applications, 2025, 45(2): 354-361.
[12]	Xintao DUAN, Mengru BAO, Yinhang WU, Chuan QIN. Active protection method for deep neural network model based on four-dimensional Chen chaotic system [J]. Journal of Computer Applications, 2025, 45(11): 3621-3631.
[13]	Rui SHI, Yong LI, Yanhan ZHU. Adversarial sample attack algorithm of modulation signal based on equalization of feature gradient [J]. Journal of Computer Applications, 2024, 44(8): 2521-2527.
[14]	Mei WANG, Xuesong SU, Jia LIU, Ruonan YIN, Shan HUANG. Time series classification method based on multi-scale cross-attention fusion in time-frequency domain [J]. Journal of Computer Applications, 2024, 44(6): 1842-1847.
[15]	Bin XIAO, Mo YANG, Min WANG, Guangyuan QIN, Huan LI. Domain generalization method of phase-frequency fusion from independent perspective [J]. Journal of Computer Applications, 2024, 44(4): 1002-1009.