PATTERN RECOGNITION USING NEURAL NETWORKS

madbrainedmudlickAI and Robotics

Oct 20, 2013 (3 years and 5 months ago)

66 views




#304,DV Arcade, Opp. Agrawala Sweets, Street No. 10, Himayath Nagar, Hyderabad
-

500 029


Ph: 040
-
3242 8143.



PATTERN RECOGNITION USING NEURAL NETWORKS

ABSTRACT


Pattern recognition is the study of how machines can observe the
environment, learn to distinguish patterns of interest from their background, and
make sound and reasonable decisions about the categories of the patterns. In this
project we use neural netwo
rks for the recognition of patterns. A computer cannot
recognize a pattern (or image) if it is stretched or compressed. In this we develop a
system which recognizes the distorted image with the original image. The unique
feature of this project is even an
image is distorted it can be recognized by the
system easily by using Neural networks. Neural Networks can be viewed as
massively parallel computing systems consisting of an extremely large number of
simple processors with many interconnections. The main c
haracteristics of neural
networks are that they have the ability to learn complex nonlinear input
-
output
relationships, use sequential procedures, and adapt themselves to the data. The
Hopfield neural network is a simple feedback neural network which is ab
le to store
patterns in a manner rather similar to the brain
-

the full pattern can be recovered if
the network is presented with only partial information. In these neural networks to
train the network we use back propagation algorithm. A Back Propagation
network
learns by example. You give the algorithm examples of what you want the network
to do and it changes the network’s weights so that, when training is finished, it will
give you the required output for a particular input. Back Propagation networks ar
e
ideal for simple Pattern Recognition and Mapping Tasks. As just mentioned, to



#304,DV Arcade, Opp. Agrawala Sweets, Street No. 10, Himayath Nagar, Hyderabad
-

500 029


Ph: 040
-
3242 8143.



train the network you need to give it examples of what you want the output you
want (called the Target) for a particular input.



Existing System:


New neural networks for
pattern recognition may be built

from existing neural networks.

An existing neural network pre
-
trained for a starting pattern is chosen based on a desired target
pattern.


Proposed Approach:

There exist several different techniques for recognizing pattern
s. One distinguishes
pattern by the number of loops in a pattern and the direction of their concavities. Another
common technique uses backpropagation in a neural network and this paper will investigate how
good neural networks solve the pattern recognitio
n problem.


Backpropagation is a technique discovered by Rumelhart, Hinton and Williams in 1986
and it is a supervised algorithm that learns by first computing the output using a feedforward
network, then calculating the error signal and propagating the er
ror backwards through the
network. For more information about the backpropagation algorithm we need to see
Introduction
to Neural Networks



S
PECIFICATION
R
EQUIRMENTS
:


Hard ware and Software Requirements:

This gives the details relating to the environment

in which the application software is
developed. Following are the hardware and software that are used in developing the application.




#304,DV Arcade, Opp. Agrawala Sweets, Street No. 10, Himayath Nagar, Hyderabad
-

500 029


Ph: 040
-
3242 8143.




Hardware Requirements:

The following are the efficient hardware requirements to run the Application




Processor


: Intel P
entium
-
IV and above



Hard disk


: 80GB Min.



RAM



: 512 MB Min. & Above



Others



: If any Applicable


Software Requirements:

The general words of software are a collection of programs and API’s each program
performs a specific task. This specification gives

a clear idea about the software can be used to
develop this project.




Front End


: C#.Net 2.0 with ASP.NET



Operating System

: Windows XP
-
2



Frame Work


: .Net 3.5 & Above