Depth-independent internal fingerprint based on optical coherence tomography

Area of Science:

• Biometrics
• Medical Imaging
• Dermatology

Background:

• Optical coherence tomography (OCT) is used for 3D fingerprint imaging but is affected by skin depth and state.
• Existing OCT methods face challenges in biometric recognition and encryption due to depth dependency.
• Developing depth-independent fingerprint features is crucial for reliable biometrics.

Purpose of the Study:

• To present a new method for extracting internal fingerprints independent of fingertip skin depth.
• To enhance biometric recognition and encryption security by overcoming surface variations.
• To utilize the epidermal-dermal junction (DEJ) for robust fingerprint feature extraction.

Main Methods:

• Segmentation of skin surface and DEJ using a deep learning algorithm.
• Extraction of internal fingerprints via maximum intensity projection (MIP) of the DEJ.
• Quantitative analysis to assess the structural similarity and accuracy of the extracted internal fingerprint.

Main Results:

• The MIP of DEJ successfully extracts internal fingerprint features.
• The internal fingerprint is independent of fingertip skin depth and surface conditions.
• The method demonstrates robustness against scar simulation and variations in skin state.

Conclusions:

• The proposed MIP of DEJ method provides a depth-independent internal fingerprint.
• This approach enhances the reliability of fingerprint-based biometric recognition and encryption.
• The technique is resilient to surface skin variations and scar simulations.