Development of a Tool for Determining the Authenticity of Signature using Raspberry Pi Board
Denyansyah H. Rohan, Muhammad Yusro* and Wisnu Djatmiko

Department of Electrical Engineering, Faculty of Engineering, Universitas Negeri Jakarta,
Jl. Rawamangun Muka, Jakarta Timur, Indonesia
*myusro[at]unj.ac.id

Abstract

This study aims to develop a signature authenticity determination tool based on image processing methods using raspberry pi. The device consists of three main parts, namely a camera module for reading signature images, a raspberry pi board for processing signature images; and an LCD module for displaying the results of signature authenticity detection.
This study proposes a simple algorithm using the binary image method and canny edge detection to change images and compare pixel values between reference signature images and doubtful signature images. This process is done by comparing pixels with the same position from different signature images. The comparison results form the basis of a decision tool for determining the authenticity of signatures. The result is displayed on the LCD screen with the name of the owner of the signature obtained from the tool database.
This system has the ability to store five signatures that serve as reference signatures and has six possible final decisions in determining the authenticity of signatures. The results showed that the system can determine the authenticity of signatures with a success rate of 81% from testing of ten different signature images.

Keywords: signature authenticity, image processing, raspberry pi

Topic: Electronics Engineering