Offline Signature Verification using Artificial Neural Network

Conformation of offline signature is one of the most difficult area for personal authentication. Throughout literature, various techniques have been studied to find out whether a given signature is certifiable or false. Signature is a critical part of making a distinction between evidence and validation of an object in this present world of digitalization. Authentication an offline signature is a crucial and worrying signature verification process. As signature can be used by many organizations and associations as a means of authentication to secure their sensitive information and also to safeguard their security because of which less information measure is accessible, which is not sufficient to identify the pattern used as a part of signature. This paper portrays a novel approach to the authentication of signatures and identifiable proof in an offline environment. In the proposed work, sample signatures are being collected in the database by scanning the actual signatures and image processing methods that will be used for generalization and to extract the features of the scanned signature. With the use of artificial neural networks such as back propagation algorithms, machine learning algorithms are used to verify signatures.