Multimodal Biometric Authentication for Enhancing Security in Online Examination Systems

View at publisher Summarise with VIRAT
Authors: Mekala N, S Subiksha, S Silpa, M Vasantha Leela, Sasirekha SP
Cited by: 0
References: 15

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

With the fast growth of online examination in higher education, the authenticity and reliability of the examination are often criticised because of impersonation, weak authentication, and ease of exploitation of traditional systems. Passwords and ID uploads are frequently abused, and unimodal biometric methods, such as facial recognition and voice recognition, are still susceptible to spoofing, poor picture quality, and distorted voices. This drawback highlights the need for a more robust and resource-efficient solution in academic networks. To mitigate this dearth, we present a novel multimodal biometric-based secure framework that jointly utilizes face and voice to verify a student during online examinations. To compute the facial and audio feature representations, a mobile-efficient facial feature extractor has been implemented to calculate the face embeddings and melfrequency cepstral coefficients (MFCC) for representing audio signals, followed by a convolutional neural network (CNN) for processing. The learned features are then fused using an adaptive weighing mechanism that combines the strengths from the two sources, ensuring the robustness of the proposed approach is guaranteed, even in the presence of low illumination or background interference. This is carried out using publicly available Kaggle datasets of faces and voices, trained on Google Colab, and a live testing interface created using Gradio. Performance evaluation on accuracy, precision, recall, F1-score, ROC-AUC, and equal error rate shows higher reliability compared to unimodal methods. The result is a practical, deployable biometric solution for undergraduates, which adds a layer of integrity and security to online exams.

Fields of Study & Subjects

Biometric Identification and Security User Authentication and Security Systems Face recognition and analysis

Publication details

Journal / source
International Conference on Cognitive Systems and Computer Interaction (ICoSCI)
Pages
1–6
Publication date
15 January 2026
Language
EN