A Hand Features Based Fusion Recognition Network with Enhancing Multi-Modal Correlation

作     者：Wei Wu Yuan Zhang Yunpeng Li Chuanyang Li YanHao 

作者机构：School of Information EngineeringShenyang UniversityShenyang110044China 

出 版 物：《Computer Modeling in Engineering & Sciences》 (工程与科学中的计算机建模（英文）)

年 卷 期：2024年第140卷第7期

页      面：537-555页

核心收录：

学科分类：12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 08[工学] 081201[工学-计算机系统结构] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：funded by the National Natural Science Foundation of China(61991413) the China Postdoctoral Science Foundation(2019M651142) the Natural Science Foundation of Liaoning Province(2021-KF-12-07) the Natural Science Foundations of Liaoning Province(2023-MS-322) 

主　　题：Biometrics multi-modal correlation deep learning feature-level fusion 

摘      要：Fusing hand-based features in multi-modal biometric recognition enhances anti-spoofing ***,it leverages inter-modal correlation to enhance recognition ***,the robustness and recognition performance of the system can be enhanced through judiciously leveraging the correlation among multimodal ***,two issues persist in multi-modal feature fusion recognition:Firstly,the enhancement of recognition performance in fusion recognition has not comprehensively considered the inter-modality correlations among distinct ***,during modal fusion,improper weight selection diminishes the salience of crucial modal features,thereby diminishing the overall recognition *** address these two issues,we introduce an enhanced DenseNet multimodal recognition network founded on feature-level *** information from the three modalities is fused akin to RGB,and the input network augments the correlation between modes through channel *** the enhanced DenseNet network,the Efficient Channel Attention Network(ECA-Net)dynamically adjusts the weight of each channel to amplify the salience of crucial information in each modal *** separable convolution markedly reduces the training parameters and further enhances the feature *** evaluations were conducted on four multimodal databases,comprising six unimodal databases,including multispectral palmprint and palm vein databases from the Chinese Academy of *** Equal Error Rates(EER)values were 0.0149%,0.0150%,0.0099%,and 0.0050%,*** comparison to other network methods for palmprint,palm vein,and finger vein fusion recognition,this approach substantially enhances recognition performance,rendering it suitable for high-security environments with practical *** experiments in this article utilized amodest sample database comprising 200 *** subsequent phase involves preparin