You are on page 1of 2

Seminer Abstract : Signature Verification

Handwritten signatures are considered as the most natural method of authenticating a

person’s identity.As signatures are the primary mechanism both for authentication and
authorization in legal transactions, the need for research in efficient auto-mated solutions for
signature recognition and verification has increased in recent years.
The basic goal of the handwritten signatures is to provide an accurate method in order to
verify a person’s identity based on the way in which he/she signs his/her name. Hence for this
reason, the handwritten signatures are widely accepted, socially and legally throughout the

Problem definition:
Extract and recognize signature from a scanned image
Input to the system is scanned image which contain signature and as the output
the system will identify the signature if match exist.

This Seminer tries to implement a signature recognition system, which will extract
signature from a scanned image and hence tries to identify the person if a match exist in the
database. Hampered by the large amount of variation between handwritten samples, researchers
have had to find techniques that will improve the ability of computers to represent and recognize
handwritten samples. One approach that has shown great impact is the use of artificial neural
network. In this approach, an artificial neural network is trained to identify similarities and
patters among different samples. This seminer will explore the topic of automated signature
recognition using artificial neural network.

Recognition is finding the identification of the signature owner. Verification is the

decision about whether the signa-ture is genuine or forgery.
Forgeries and its types:
There are three types of forgeries:
1. Random Forgery - In random forgery, the person doesn’t have the shape of the original
signature. The signer uses the name of the victim in his own style to create a forgery known as
the simple forgery or random forgery.
2. Unskilled Forgery - The signer initiates the signature in his own style without any
knowledge of the spelling and does not have any prior experience.
3. Skilled Forgery – undoubtedly professional impostors or persons who have experience in
copying the
signature create the most difficult of all forgeries.

In this paper we present an off-line signature verification and recognition system using
the global, directional and grid fea-tures of signatures. Support Vector Machine (SVM) was used
to verify and classify the signatures and a classification ratio of 0.95 was obtained. As the
recognition of signatures repre-sents a multiclass problem SVM's one-against-all method was
used. We also compare our methods performance with Artifical Neural Network’s (ANN)
backpropagation method

SRVS (Signature Recognition and Verification System) is often categorized in two

major classes: on-line SRVS and off-line SRVS. The difference of on-line and off-line lies in
how data are obtained. In the on-line SRVS data are obtained using an electronic tablet and other
devices. In the off-line SRVS images of the signatures written on a paper are ob-tained using a
scanner or a camera