Personal Authentication Mechanism Based on Finger Knuckle Print

Area of Science:

• Biometrics and Authentication
• Pattern Recognition
• Computer Vision

Background:

• Biometric traits like fingerprints are widely used for user identification and verification.
• Hand-based biometrics, including vein and finger knuckle patterns, are emerging as important authentication modalities.
• Existing biometric systems face challenges in balancing security and user convenience.

Purpose of the Study:

• To propose a novel methodology for secure biometric authentication utilizing Finger Knuckle Print (FKP) patterns.
• To enhance the performance of hand-based biometric systems by reducing the False Rejection Rate (FRR) without compromising the False Acceptance Rate (FAR).
• To evaluate the effectiveness of the proposed FKP authentication system compared to conventional methods.

Main Methods:

• Extraction of texture patterns from finger knuckle images using the Gabor filter combined with the Expectation-Maximization (EM) algorithm.
• Acquisition of feature vectors from extracted texture patterns using the Scale Invariant Feature Transform (SIFT) algorithm.
• Performance evaluation based on Genuine Acceptance Rate (GAR) and False Rejection Rate (FRR).

Main Results:

• The proposed FKP authentication methodology demonstrates improved performance over conventional hand-based modalities.
• The system effectively reduces the False Rejection Rate (FRR) while maintaining a low False Acceptance Rate (FAR).
• The user-friendliness of data collection for FKP biometrics is highlighted as a significant advantage.

Conclusions:

• Finger Knuckle Print (FKP) recognition offers a secure and efficient biometric authentication solution.
• The combination of Gabor filters, EM algorithm, and SIFT provides robust feature extraction for FKP.
• FKP authentication presents a promising alternative for secure and user-friendly identification systems.