Deep Feature Fusion of Fingerprint and Online Signature for Multimodal Biometrics

Leghari, Mehjabeen; Memon, Saima Q.; Dhomeja, Lachhman Das; Jalbani, Akhtar Hussain; Chandio, Asghar Ali

doi:10.3390/computers10020021

Cited by 27 publications

(17 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Leghari et al [25] compared feature-level fusion strategies on multimodal biometric data and showed that higher accuracy was obtained by performing fusion at the convolutional layer than at the linear layer. Selective fusion networks [26] weight the estimated high-quality information with the original depth information as a whole, but ignore the intra-modal interchannel variability.…”

Section: A Multimodal Biometric Recognitionmentioning

confidence: 99%

Multimodal Finger Recognition Based on Asymmetric Networks With Fused Similarity

Huang

Wang

2023

IEEE Access

View full text Add to dashboard Cite

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-toend multimodal finger recognition model that incorporates attention mechanisms into a similarity-aware encoder to achieve accurate recognition results for accurate identification 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.INDEX TERMS Multimodal biometric recognition, feature fusion, autoencoder, deep learning.

show abstract

Section: A Multimodal Biometric Recognitionmentioning

confidence: 99%

Multimodal Finger Recognition Based on Asymmetric Networks With Fused Similarity

Huang

Wang

2023

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Tiong et al [45] proposed a method of information fusion via extracting features from raw biometric data using a CNN and then combining them with a series of fully connected layers. Other deep learning approaches have been proposed recently [7,53,2,28,44]. Contrasting these methods, we opt for a linear projectionbased approach to limit the multiplicative depth of the circuit and decrease computational complexity, which is important for creating a practical solution in FHE.…”

Section: Related Workmentioning

confidence: 99%

HEFT: Homomorphically Encrypted Fusion of Biometric Templates

Sperling¹,

Ratha²,

Ross³

et al. 2022

Preprint

View full text Add to dashboard Cite

This paper proposes a non-interactive end-to-end solution for secure fusion and matching of biometric templates using fully homomorphic encryption (FHE). Given a pair of encrypted feature vectors, we perform the following ciphertext operations, i) feature concatenation, ii) fusion and dimensionality reduction through a learned linear projection, iii) scale normalization to unit 2 -norm, and iv) match score computation. Our method, dubbed HEFT (Homomorphically Encrypted Fusion of biometric Templates), is customdesigned to overcome the unique constraint imposed by FHE, namely the lack of support for non-arithmetic operations. From an inference perspective, we systematically explore different data packing schemes for computationally efficient linear projection and introduce a polynomial approximation for scale normalization. From a training perspective, we introduce an FHE-aware algorithm for learning the linear projection matrix to mitigate errors induced by approximate normalization. Experimental evaluation for template fusion and matching of face and voice biometrics shows that HEFT (i) improves biometric verification performance by 11.07% and 9.58% AUROC compared to the respective unibiometric representations while compressing the feature vectors by a factor of 16 (512D to 32D), and (ii) fuses a pair of encrypted feature vectors and computes its match score against a gallery of size 1024 in 884 ms. Code and data are available at https://github.com/humananalysis/encrypted-biometric-fusion

show abstract

“…Leghari et al [38] discussed the fusion of fingerprint and online signature. The CNN model was developed for early and late feature fusion, and in the early feature fusion obtained an accuracy of 99.1%, and in the late feature fusion obtained an accuracy of 98.35%.…”

Section: Et Al [32] Used Incompletely Closed Nir (Icnir) Finger Capturing Equipment Based On Nir Imaging Technique To Perform Feature-levmentioning

confidence: 99%

A novel multimodal hand database for biometric authentication

2022

IJATEE

View full text Add to dashboard Cite

Automated person identification through a biometric device is the most popular technology for individual authentication. Recognition is based on the behavioral or physical characteristics such as face, fingerprints, voice, stride, and signature [1]. These characteristics are known as biometric cues or biometric modalities. For the past several years, these biometrics cues have been used in applications ranging from less security to high security [2−5].Due to the defensive nature of multimodal biometrics in terms of security and robustness, it is preferred over unimodal biometric systems [6].

show abstract

Deep Feature Fusion of Fingerprint and Online Signature for Multimodal Biometrics

Cited by 27 publications

References 33 publications

Multimodal Finger Recognition Based on Asymmetric Networks With Fused Similarity

Multimodal Finger Recognition Based on Asymmetric Networks With Fused Similarity

HEFT: Homomorphically Encrypted Fusion of Biometric Templates

A novel multimodal hand database for biometric authentication

Contact Info

Product

Resources

About