Identifying fingerprint spoofs using skin properties

Optical fingerprint sensors enable seamless authentication and bezel-less displays by directly integrating a fingerprint sensor under the display screen of a device such as a smartphone. While conventional optical fingerprint sensors provide a convenient authentication technique, such sensors can be insecure due to the ease of generating and using spoofs that pass the authentication test. This disclosure describes techniques that utilize various skin properties to detect spoofs. Non-uniform illumination from the sensor interacts with the 3D structure of the finger and produces brightness variation that can be utilized to detect spoofs. Further, skin properties such as ridge-valley distance, skin deformation under pressure, and macro properties such as the shape of a finger can also be utilized to detect spoofs. Spoof detection can be performed to reject 2D or 3D spoofs prior to performing fingerprint matching