Next-Generation Biometric Authentication: Overcoming the Twin Identification Challenge with Advanced Facial Recognition and Multi-Modal Analysis Techniques
Conventional facial recognition struggles with individuals sharing near-identical facial features, particularly monozygotic twins. This research introduces a robust, real-time methodology to overcome this by integrating geometric, textural, and dynamic facial characteristics. The framework employs M...
Saved in:
| Published in: | Journal of Al-Qadisiyah for Computer Science and Mathematics Vol. 17; no. 3 |
|---|---|
| Main Authors: | , |
| Format: | Journal Article |
| Language: | English |
| Published: |
30.09.2025
|
| ISSN: | 2074-0204, 2521-3504 |
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Conventional facial recognition struggles with individuals sharing near-identical facial features, particularly monozygotic twins. This research introduces a robust, real-time methodology to overcome this by integrating geometric, textural, and dynamic facial characteristics. The framework employs Multi-Task Cascaded Convolutional Networks (MTCNN) for face detection and alignment, followed by FaceNet for 128-dimensional facial embedding generation. MediaPipe's 468-point facial landmark extraction quantifies subtle structural variations via transformation matrix analysis and blend-shape evaluation, capturing static geometric discrepancies and dynamic micro-expressions. Validated on 7,200-image dataset (70% training, 30% testing), the system achieved 97.73% accuracy, operating efficiently on consumer-grade GPUs. This approach significantly enhances biometric technology, offering improved identity verification for genetically similar individuals in critical security applications like border control and secure access management, thereby addressing a key limitation in current facial recognition systems. |
|---|---|
| ISSN: | 2074-0204 2521-3504 |
| DOI: | 10.29304/jqcsm.2025.17.32383 |