Residual learning

Abstract—Fingerprint recognition technology has become

popular for mobile device authentication systems due to its

reliability and ease of use. As smartphones evolve, fingerprint

sensors are now integrated into smartphone power with a width

of 2.2 mm. However, tiny sensor sizes have led to limited finger

coverage and external factors such as sweat or water droplets

can cause image distortion, making user authentication more

challenging.

To address these issues, we propose the FFP-UNet, which uses