Secure Human Identification with Multi-Modal Biometrics Using Weighted Jackal Optimization and Transformer-Based Squeezenet
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
The restrictions of unimodal systems have resulted in the introduction of multimodal biometric systems. The paper introduces a skeleton based fusion-based multimodal identification system which involves the use of fingerprint, palmprint and iris data. To start with, a guided non-local filter enhances the quality of the image whereas SiaNet is used to extract discriminative features. Parallel maximum covariance fusion then fuses multimodal embeddings, optimizing the best features using a jackal optimization strategy with weights. Lastly, a transformer based SqueezeNet classifier is used to perform recognition. Its system has 99.64 percent accuracy, 93.33 percent and 97.76 percent 5 and 10 fold cross-validation accuracy respectively. It is a computationally simple model which can be applied in real time identification tasks.
Fields of Study & Subjects
Biometric Identification and Security
Gait Recognition and Analysis
User Authentication and Security Systems
Publication details
- Journal / source
- International Conference on Intelligent Systems for Communication, IoT and Security (ICISCoIS)
- Pages
- 1–6
- Publication date
- 8 January 2026
- Language
- EN