NYU Tandon researchers create synthetic fingerprints capable of spoofing smartphone fingerprint sensors. Devices typically allow users to enroll several different finger images, and a match for any saved partial print is enough to confirm identity. Frankly, that is not how I envisaged these biometrics scans to work.  Partial fingerprints are less likely to be unique than full prints, and research demonstrated that enough similarities exist between partial prints to create MasterPrints capable of matching many stored partials in a database.  Read more below.

Source: Machine Learning Masters the Fingerprint to Fool Biometric Systems