Multimodal Finger Recognition Based on Asymmetric Networks With Fused Similarity
Multimodal biometric system has received increasing interest as it offers a more secure and accurate authentication solution than unimodal systems. However, existing biometric fusion methods are still inadequate in dealing with correlations and redundancy of multimodal features simultaneously, causi...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10038659/ |
| _version_ | 1797894929862623232 |
|---|---|
| author | Yiwei Huang Hui Ma Mingyang Wang |
| author_facet | Yiwei Huang Hui Ma Mingyang Wang |
| author_sort | Yiwei Huang |
| collection | DOAJ |
| description | Multimodal biometric system has received increasing interest as it offers a more secure and accurate authentication solution than unimodal systems. However, existing biometric fusion methods are still inadequate in dealing with correlations and redundancy of multimodal features simultaneously, causing bottlenecks in performance improvement. To overcome the above problem, this paper proposes an end-to-end multimodal finger recognition model that incorporates attention mechanisms into a similarity-aware encoder for accurate recognition results. Firstly, due to the different distribution of fingerprint and finger vein images, we propose a finger asymmetric backbone network (FAB-Net) for extracting discriminative intra-modal features, which reduces the network width by efficient utilization of feature maps. Then, a novel attention-based encoder fusion network (AEF-Net) with fused similarity performs dimensionality reduction-based fusion on multimodal multilevel features to alleviate performance degradation due to information redundancy. We also introduce channel attention in AEF-Net, which differs from the traditional attention mechanism by considering interdependencies between modalities to further improve performance. Extensive recognition experiments are conducted on three multimodal finger databases to verify the effectiveness of our method compared to state-of-the-art methods. Detailed ablation studies have also been carried out, which demonstrated that encoder-based reconstruction of redundant information can improve recognition performance. |
| first_indexed | 2024-04-10T07:17:09Z |
| format | Article |
| id | doaj.art-40c54db499f24353a2c42a26b088bec1 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-10T07:17:09Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-40c54db499f24353a2c42a26b088bec12023-02-25T00:02:54ZengIEEEIEEE Access2169-35362023-01-0111174971750910.1109/ACCESS.2023.324298410038659Multimodal Finger Recognition Based on Asymmetric Networks With Fused SimilarityYiwei Huang0Hui Ma1https://orcid.org/0000-0002-0513-2462Mingyang Wang2College of Electronic Engineering, Heilongjiang University, Harbin, ChinaCollege of Electronic Engineering, Heilongjiang University, Harbin, ChinaCollege of Electronic Engineering, Heilongjiang University, Harbin, ChinaMultimodal biometric system has received increasing interest as it offers a more secure and accurate authentication solution than unimodal systems. However, existing biometric fusion methods are still inadequate in dealing with correlations and redundancy of multimodal features simultaneously, causing bottlenecks in performance improvement. To overcome the above problem, this paper proposes an end-to-end multimodal finger recognition model that incorporates attention mechanisms into a similarity-aware encoder for accurate recognition results. Firstly, due to the different distribution of fingerprint and finger vein images, we propose a finger asymmetric backbone network (FAB-Net) for extracting discriminative intra-modal features, which reduces the network width by efficient utilization of feature maps. Then, a novel attention-based encoder fusion network (AEF-Net) with fused similarity performs dimensionality reduction-based fusion on multimodal multilevel features to alleviate performance degradation due to information redundancy. We also introduce channel attention in AEF-Net, which differs from the traditional attention mechanism by considering interdependencies between modalities to further improve performance. Extensive recognition experiments are conducted on three multimodal finger databases to verify the effectiveness of our method compared to state-of-the-art methods. Detailed ablation studies have also been carried out, which demonstrated that encoder-based reconstruction of redundant information can improve recognition performance.https://ieeexplore.ieee.org/document/10038659/Multimodal biometric recognitionfeature fusionautoencoderdeep learning |
| spellingShingle | Yiwei Huang Hui Ma Mingyang Wang Multimodal Finger Recognition Based on Asymmetric Networks With Fused Similarity IEEE Access Multimodal biometric recognition feature fusion autoencoder deep learning |
| title | Multimodal Finger Recognition Based on Asymmetric Networks With Fused Similarity |
| title_full | Multimodal Finger Recognition Based on Asymmetric Networks With Fused Similarity |
| title_fullStr | Multimodal Finger Recognition Based on Asymmetric Networks With Fused Similarity |
| title_full_unstemmed | Multimodal Finger Recognition Based on Asymmetric Networks With Fused Similarity |
| title_short | Multimodal Finger Recognition Based on Asymmetric Networks With Fused Similarity |
| title_sort | multimodal finger recognition based on asymmetric networks with fused similarity |
| topic | Multimodal biometric recognition feature fusion autoencoder deep learning |
| url | https://ieeexplore.ieee.org/document/10038659/ |
| work_keys_str_mv | AT yiweihuang multimodalfingerrecognitionbasedonasymmetricnetworkswithfusedsimilarity AT huima multimodalfingerrecognitionbasedonasymmetricnetworkswithfusedsimilarity AT mingyangwang multimodalfingerrecognitionbasedonasymmetricnetworkswithfusedsimilarity |