Off-line signature verification system using Bayesian regularization backpropagation

Mok, Zee Juin (2017) Off-line signature verification system using Bayesian regularization backpropagation. Project Report. Universiti Teknikal Malaysia Melaka, Melaka, Malaysia. (Submitted)

Text (Full text) Off-line signature verification system using Bayesian regularization backpropagation.pdf - Submitted Version Download (15MB)

Abstract

The signature verification is been used to verify a person carrying out daily routine procedures, like banking operations, document sigining and online fee transfer. Signature is can be easily forged by skilled personnel, an automated signature verification method should be developed as human peer verification is not accurate. This project dealing with an automated method of verification an off-line signature recognition by extracting features that characterize the signature. The approach starts by scanning or collecting data into the computer using scanner or drawing pad, then process the quality of images through enhancement and noise reduction, followed by feature extraction and neural network training, and finally verifies whether a signature is original or fake. This system can be used by bank and other activity that related to a handwritten signature, this allows them to analyze and verify their customer signature originality to prevent bad intention personal to steal important information.

Actions (login required)

View Item