Abstract
Signature verification is one in every of the foremost wide used biometrics for authentication. The target of the signature verification system is to discriminate between two classes: the initial and also the forgery, that are associated with intrapersonal and social variability. Firstly, there exists great variation even between two signatures of identical person. They never begin from identical position and neither do they terminate at constant position. Also, the angle of inclination of the signatures, the relative spacing between letters of the signatures, height of letters, all vary even for constant person. Therefore it becomes a difficult task to check between two signatures of identical person. The proposed an offline signature verification system to require care of that, that relies on depth for segmentation of signature image into completely different elements, at that time geometric center of the every section is locate out because the feature purpose of that section.