The influence of fingerprint image degradations on the performance of biometric system and quality assessment
Résumé
One of the main challenges facing biometric technologies is system performance decreasing
caused by low quality biometric samples. In fingerprint recognition, the system performance may be negatively affected by fingerprint image degradations, which are introduced by subject characteristic, image acquisition, subject behavior, or environment. Therefore, it is necessary to investigate how different fingerprint image degradations influence biometric system performance. In this paper, we will first study different fingerprint image degradations that affect system performance. Then review state-of-the-art fingerprint sample quality assessment methods and their evaluation approaches. Based on the survey, we select corresponding degradations and apply them to fingerprint samples. The system performance comparison between original and degraded fingerprints will be conducted in order to illustrated the impact of each degradation on biometric system performance. Finally, we use NFIQ fingerprint image quality metric to investigate its performance on selected degradations.