Offline Signature Verification based on Pixel Oriented and Component Oriented Feature Extraction

Abstract

Signature verification is one of the most widely used biometrics for authentication. The proposed technique is based on pixel oriented and component oriented features extraction. The aim of the proposed work is to develop an automatic signature verification system which takes the input signature of the known user and either accepts or rejects it after extracting and comparing the different pixel oriented and component oriented features of the input signature with the already trained features of the reference signature.This technique is suitable for various applications such as bank transactions, passports etc. The threshold used in the proposed technique can be dynamically changed according to the target application. Basically, the threshold here is the security level which the user can input as per his requirement. The proposed technique deals with skilled forgeries and has been tested on two databases: Database A and a standard Database B (Set 1 and Set 2). The proposed technique gives FAR of 9.63% and FRR of 8.46% for Database A, FAR of 12.21% and FRR of 10.40% for Database B (Set 1) and FAR of 15.87% and FRR of 13.72% for Database B (Set 2) which is better than many existing verification techniques.

Keywords: Offline Signature Verification, Skilled Forgery, Grid Feature Extraction, False Acceptance Rate (FAR), False Rejection Rate (FRR), Pixel oriented features, Component oriented features.