Combined Static and Dynamic Features Extraction from Handwritten Signature

Abstract

Handwritten signatures are a vital part of everyone’s life. They are widely used for authentication. The biggest challenge in this technology world is the security of our signatures stored within the bank and other government organizations. All organisations always live in fear of losing their data, which would be exposed and could be accessed by an intruder and used without authorization. The best way to secure signatures is to store them in encrypted form using symmetric or asymmetric encryption approaches. Both methods depend on a unique key that is used for encryption and decryption. But storing and remembering the key is another problem. If this key gets into the hands of an unauthorised user, he can access all the confidential data. Another way to secure signatures without storing and remembering the key is to use biometric entities themselves, such as a handwritten signature, fingerprint, iris, etc. These biometric entities are unique by nature and always remain within a user; they cannot be stolen or exposed to others. In this paper, we defined the key(s) that are derived from the handwritten signature to encrypt the signature itself. These keys are nothing but the features associated with the signature extracted from online as well as offline signatures. Features of a handwritten signature are used during the verification of an original or forged signature as well as the security of it. This paper defines the keys that are derived from the online and offline features of the handwritten signature to encrypt the signature itself.