ML-based fingerprint compression at higher rates: balancing visual fidelity and biometric accuracy

Key Points

• The aim is to investigate how machine learning can compress fingerprints efficiently while maintaining quality and accuracy.
• Developed a machine learning algorithm for fingerprint compression.
• Tested the algorithm at various compression rates.
• Evaluated visual fidelity and biometric accuracy through performance metrics.
• Achieved significant compression rates while preserving high visual fidelity.
• Biometric accuracy remains above industry standards at higher data rates.
• Demonstrated a trade-off between compression and fidelity that can be optimized.