Fingerprint identification based on semi-supervised FSS

摘要: Fingerprint images captured in real world applications always include some variations, called intra-class due to various uncontrolled conditions like scratching, aging, moisting, drying, etc. It is important for current fingerprint identification systems adaptively deal with these variations. In this paper, we propose a semi-supervised FSS based method. We use unlabeled samples train Center each finger setting, which significantly improves the robustness of evaluate our method on DIEE database. The experimental results show favorable performance as compared state-of-the-art.