Customizable Presentation Attack Detection for Improved Resilience of Biometric Applications Using Near-Infrared Skin...

Area of Science:

• Biometrics
• Security Engineering
• Optical Sensing

Background:

• Biometric systems are crucial for digital identity management, demanding robust security against presentation attacks (e.g., fake fingerprints, masks).
• Existing biometric modalities are vulnerable to spoofing attempts using artificial artifacts.
• Near-infrared (NIR) technology offers a contactless method to differentiate human skin from artifact materials.

Purpose of the Study:

• To introduce a novel, miniaturized near-infrared presentation attack detection (NIR-PAD) device.
• To enhance the reliability and resilience of biometric authentication systems.
• To present a versatile platform for sensor fusion and integration into future biometric systems.

Main Methods:

• Development of a miniaturized NIR-PAD device.
• Implementation of an innovative signal processing chain.
• Integration of a distance measurement feature for enhanced accuracy.

Main Results:

• The NIR-PAD device successfully distinguishes between human skin and common artifact materials.
• The integrated distance measurement boosts system reliability and resilience.
• The modular design facilitates adaptable integration into various biometric scanners.

Conclusions:

• The developed NIR-PAD device represents a significant advancement in combating biometric spoofing.
• Its adaptable and integrated design makes it a promising solution for enhancing security in diverse biometric applications.
• Future work may focus on further enhancements and broader application of this technology.