Abstract
This paper presents a robust signature verification and forgery detection system using Additive fuzzy and TS modeling technique. The features of various handwritten signatures are sampled with proper analysis and encapsulated to devise an effective verification system. Grid method was used to extract features angles for detection of forgeries and verification of genuine signatures. In financial Accounting, Auditing and Forensic Investigation, signature forgery could occur in various ways. This could be carried out on papers, sales documents such as invoices or inventory procurement requisition paper, title documents on landed property or other tangible assets. It is also perpetrated on payment authorization such as cheques, payment vouchers both in cash and on bills. During this exercise, the fraud perpetrators perfect their concentration on the surface paper, and trace the original signature from the mandate given earlier. It has been difficult to use accounting and auditing professions to track down financial fraud in Nigeria mostly with the problem of unearthing ingenious fraud. Exponential membership function was used to fuzzified the derived functions, and modified into structural parameters suitable to adapt to any possible variations that may result from handwriting styles and also to reflect any other factors due to scripting of a signature. The proposed system is tested on a large database of signatures obtained from 40 subjects.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have