Published on Nov 30, 2023
Dynamic Signature Verification Using Pattern Recognition Project deals with knowledge base creation where sample signatures and twenty-five pixel codes for characters and numbers are stored.
The second and third phase involves character recognition (using perceptron and BPN) and number recognition (using perceptron and seven segment) wherein the user is made to enter twenty-five pixel code for each character or number. The output is recognized character or number of input. The final phase of project deals with signature verification. The files containing signatures are given as inputs. They are compared and result is generated.
The aim of the proposed system is to recognize any alphanumeric characters. Here the alpha numerals are represented as patterns. The secondary objective of this system is Signature Verification using Sobel Edge Detection algorithm. This system can be used in lots of applications where they maintain computer-coding sheets and accept input from user like entrance exam forms, banking etc.
The system helps users to recognize hand written characters and numbers. It also allows the users to compare two signatures. Twenty-five pixel values for each hand written character or number is entered in the system where each value is either zero or one (binary inputs). In perceptron, the input is compared with stored values and using the activation function output is displayed. In BPN, the input is used to train the network. The generated result is compared with stored result and output is displayed. The sample signatures are scanned and stored as JPG file. The input signature is then scanned and stored as JPG file. The signatures are compared and results are displayed.
The minimum hardware required for the development of the project is:
Processor Type : Pentium -IV
Speed : 2.4 GHZ
Ram : 128 MB RAM
Hard disk : 20 GB HD
Operating System : Win2000/XP
Programming Package : JAVA
Tools : eclipse
SDK : JDK 1.5.0
Logical Flow
This project aimed at Dynamic Signature Verification using Pattern Recognition has satisfied the goal. The development and implementation of this system has given us a great satisfaction. Through our efforts, we have incorporated into the system several features such as flexibility and robustness. The implementation of the project provides the following: User-friendly interface Reusability of code Extensibility –Other neural networks can added Scalability – Further features can be easily added
