Score-Level Fusion of 3D Face and 3D Ear for Multimodal Biometric Human Recognition

Area of Science:

• Biometrics
• Computer Vision
• Pattern Recognition

Background:

• Unimodal biometric systems suffer from limitations like occlusion and illumination variations.
• Existing 2D biometric methods struggle with environmental factors affecting recognition accuracy.
• Multimodal biometrics offer enhanced robustness and accuracy by combining multiple traits.

Purpose of the Study:

• To propose a novel multimodal biometric system for human recognition.
• To leverage three-dimensional (3D) face and ear data for improved identification.
• To overcome the limitations of unimodal and 2D biometric systems.

Main Methods:

• Utilized Principal Component Analysis (PCA) for 3D face recognition.
• Employed Iterative Closest Point (ICP) algorithm for 3D ear recognition.
• Implemented score-level fusion to integrate 3D face and ear recognition results.

Main Results:

• The proposed multimodal system achieved a high accuracy of 99.25%.
• Score-level fusion of 3D face and ear data significantly enhanced recognition performance.
• Comparative analysis demonstrated superior accuracy compared to state-of-the-art biometric algorithms.

Conclusions:

• The proposed 3D multimodal biometric system effectively enhances human recognition accuracy and robustness.
• Score-level fusion is a viable strategy for integrating multiple biometric modalities.
• This approach offers a promising solution for secure and reliable identity verification.