Multi-Biometric Fusion of Iris and Face Dynamics using Deep Learning for High-Security Human Identification

Enhancing Biometric Security Using Artificial Neural Network-Based Multimodal Fusion of Facial Recognition and Fingerprint Identification

Biometric authentication systems based on a single modality remain vulnerable to spoofing, acquisition noise, and intra-class variation, limiting their reliability in high-security access-control applications. This study presents an artificial neural network-based multimodal framework that combines facial recognition and fingerprint identification to improve authentication accuracy, robustness, and presentation-attack resistance. Facial features are extracted using a convolutional neural network, while fingerprint texture and minutiae representations are obtained using Gabor filters and a denoising autoencoder. Feature-level and score-level information is integrated through a multilayer-perceptron meta-learner, followed by an adaptive decision module incorporating modality-specific liveness assessment. The framework was evaluated using the Labeled Faces in the Wild dataset, FVC2006, and the custom Bayelsa Multimodal Biometric Dataset comprising 320 subjects. On the reported BMBD test split, the system achieved 99.14% verification accuracy, a false acceptance rate of 0.12%, a false rejection rate of 0.34%, and an equal error rate of 0.19%. The reported performance exceeded the best unimodal baseline by 6.8 percentage points and the strongest compared multimodal method by 1.93 percentage points. End-to-end inference required 143 ms on the stated embedded platform. The ablation results indicated that learned fusion, metric-learning losses, and liveness detection each contributed to performance. These findings support the feasibility of the proposed framework under the reported experimental conditions, while broader independent evaluation remains necessary.

Robust User Authentication Using CNN-Based Face, Fingerprint, and Iris Biometrics

User authentication plays an essential role in security assurance in the present-day digital world. Traditional password-based authentication approaches are increasingly becoming vulnerable to attacks, while single modal biometric systems suffer from challenges such as noise susceptibility, deception, and intra-class variations. This paper proposes a secure and reliable multimodal biometric system using fingerprints, facial characteristics, and irises using Convolutional Neural Network (CNN) to address these weaknesses. In this system, self-learning CNN based representations are utilized for every biometric characteristic, allowing for automatic extraction of distinct deep representations without hand-crafted features. The obtained confidence or attributes across different biometrics are merged to create a combined representation for authentication. Confidence-based classification approach is utilized for authenticating the genuine and impersonator user. Efficiency of the proposed system has been tested using unique assessment criteria including accuracy, recall, true negative rate, false positive rate, precision, and false negative rates. Experimental outcomes demonstrate that the presented CNN-based multimodal biometric system achieves superior accuracy and robustness compared to the conventional single-modal systems and thus it can be used for practical security applications where reliability is required.

Multimodal Biometric Recognition Based on Face and Video Via Vision Attention Fusion Transformer

Biometric recognition systems have become an indispensable piece of contemporary security and recognition administration frameworks. Multimodal biometric systems incorporate multiple biometric features for verification provide significant advantages over uni-modal biometric systems, to name a few being, improved security, enhanced accuracy and considerable amount of robustness. Across a sample of individual modalities, face recognition is extensively applied for its reliability and ease of use, voice recognition is eminent for its high accuracy and stability. This study highlights the significance of multimodal biometric authentication method using face images and voice signals called, Multi Task Cascaded Pre-emphasis and Vision Attention Fusion Transformer (MTCP-VAFT) to enhance the security of existing multimodal biometric recognition systems in indoor surveillance videos. It introduces a novel pre-processing model that simultaneously processes face images using Multi Task Cascaded Neural model and voice signals utilizing and Pre-emphasis Filter. This work also proposes a novel Vision Attention Fusion Transformer based multimodal biometric authentication to exploit their complementary advantages and mitigate attacks. Finally the method incorporates the two modalities via an attention fusion mechanism to highlight the multimodal biometric recognition system reliability under varying circumstances. Experimental results on the MSU-AVIS dataset demonstrate the efficiency of our method, showing notable improvements achieving overall multimodal biometric recognition accuracy by 97% and improving the peak signal to noise ratio by 48.25dB. These findings demonstrate that modality-aware fusion using MTCP-VAFT method can delivery secure and flexible biometric authentication suitable for deployment on high-security platforms.

Cancelable Multi-Biometric Identification using Incremental Deep Learning for Secure and Scalable Face-Iris Recognition

The widespread use of biometric identification systems has raised important issues regarding template security and scalability. In contrast to passwords, biometric characteristics, once exposed, cannot be withdrawn, thus raising permanent risks of identity exposure. Moreover, traditional systems require complete retraining when adding new users, thus causing computational inefficiency and low scalability. This work presents an original framework for cancelable multi-biometric identification that combines face and iris dynamics in an incremental deep learning model. The proposed method produces non-invertible and revocable templates through deep feature extraction by ResNet-50, followed by a random projection transformation, thus meeting the ISO/IEC 24745:2022 security requirements. A dynamic 1D-CNN model, enhanced by elastic weight consolidation and rehearsal learning, enables incremental learning without catastrophic forgetting or complete model retraining. Experimental evaluation shows an average recognition rate of 98.98% on incremental datasets, with false acceptance rates as low as 0.115% and true acceptance rates of 98.93%. The proposed framework provides an optimal trade-off among security, scalability, and recognition performance, thus filling the most important gaps in current biometric systems and offering a basis for the development of a next-generation privacy-preserving identification infrastructure.

Toward High Accuracy and Strong Security: Cancellable Templates for Multimodal Biometric Recognition Based on Feature Fusion

Multimodal biometric recognition integrates complementary information from different modalities and significantly improves recognition accuracy and system security compared with unimodal methods. However, with the widespread adoption of these systems, the risks of biometric template leakage and theft have emerged, posing serious threats to user privacy. To address this issue, we propose a secure multimodal template protection framework based on palmprint and palmvein. We design a feature-level fusion network, PalmSynNet, where the Multi-Scale Local Feature Extraction module enhances the extraction of local features, and the Global Feature Fusion module enables cross-modal global feature interaction, resulting in robust and highly discriminative fused representations. In addition, we propose a novel cancellable random projection method, which generates protected templates through SoftMax-based random projection (SoRP) combined with other hashing algorithms, effectively avoiding the reversibility problem of random projection under certain conditions. Extensive evaluations conducted on three multimodal palmprint databases—PolyU, TJ, and CUMT—demonstrate that proposed framework not only achieves superior fusion recognition performance compared with state-of-the-art methods, but also satisfies essential security requirements, including irreversibility, revocability, unlinkability, and resistance to various attacks.

A Secure Multimodal Biometric Authentication Framework Using Watermarking and Feature-Level Fusion for Cyber-Physical System