New features extraction and application based on Gaussian-Hermite moments in fingerprint classification

Fingerprint classification is an important stage in automatic fingerprint identification systems (AFIS). Key to this process is feature extraction. For fingerprint images, there are two special features singular points (SPs) - core and delta points. Most current classification methods, no matter what they are. structural methods or network-based methods, are based on the extraction of such singular points. In this paper, we propose a new algorithm for the features extraction in Fingerprint, which is based on the distribution of Gaussian-Hermite moments of different orders in the fingerprint image. Unlike the common method, we classify the singular points into three types. With these features. we propose a method for fingerprint classification. This method has been tested on the NIST special fingerprint database 4. For the 4000 images in this database, the classification accuracy reaches 87.2 % for the 5-class problem.