Replicating the signature of a person other than the one who owe, called a signature forgery. Generally, Artificial Intelligence and signal processing techniques employed in generating such audios called signature forgery .
According to a Daily mail article on 18th February 2018, Royal Bank of Scotland (RBS) officially apologized to a customer named Jean Mackay, and paid a compensation fine of 500 pounds, for charging a fee for the Pension Protection Fund (PPF) while the client was oblivious to the Enrollment. Later investigations revealed that a mysterious culprit came up with fraudulent signatures of the victim and the analysis showed them not a match at all. Moreover, RBS admits forging an elderly customer’s signature, which left her signed up for a product she didn’t want.
This just one of many instances where forged signatures have led to a venal skullduggery. The issue isn’t localized to a particular region, but is persistently ubiquitous. This dilemma urges researchers to come up with a solution to solve the problem of impermissible signature replication. An automated approach to verify the signature becomes all the more crucial under the given circumstances.
A signature forgery composed of special characters and flourishes and uses to authentication one human beings from another. In this authentication process a captured signature stored in a computer in the form of image file. The problem is to compare the user signature with a sample database signature. The distinctiveness of a handwritten signature helps to prove the identity of the signer, while the act of signing a document represents the signer’s acceptance of, its terms and also codifies the document’s contents as being official and complete at the time it was signed.
The aim is to come up with a solution to the dilemma of signature replication. Various automated methodologies involving deep and machine learning techniques of Artificial Intelligence employed. Prior works anatomized and better solutions promulgated for the overall betterment of the system’s functioning.
Kinds of forgeries
The forgeries involved in handwritten signatures categorized based on their characteristic features. The various kinds of forgeries into the following types:
- Random forgery—the signer uses the name of the victim in his own way to make a forgery known as the random forgery or simple forgery. Signature forgery accounts for the majority of the venal skulduggery although they are rather easy to detect by un-aided eye.
- Unskilled forgery—the signer emulates the signature in his personal style with neither prior knowledge of the chronology of alphabets nor any experience. The imitation preceded by perceiving the signature with a bird’s eye view.
- Skilled forgery—undoubtedly the most treacherous form of all forgeries generated by professional impostors or people with much prior experience in emulating signatures. For achieving this one could either trace or imitate the signature by hard way. The proposed system involves Natural Language Processing (NLP) techniques that figure out the intrinsic nature of the strokes and intensity of impression provided with the appropriate tolerance to comprehend the basic stylometry for a user. Afterwards, similar traits figured out for the target image. Different categories of forgery arise depending on what limit of variation we allow over the inherent dissimilitude.
Applications of signature forgery
- Authorized checking system in banks.
- Degree Verification systems.
- Judicial cases such as decision of heritage will.