WELCOME

A Neural Network For

Handwritten Digit
Recognition
S.Praveen Kumar
(07Q71A0597)
&
V.Tharun Reddy
(07Q71A05B8)

Final Year
Dept. of Computer Science & Engineering
Avanthi Engineering College
 AGENDA
 How and Where Handwritten Digit Recognition is
used?

 What are the several techniques used for

Preprocessing the Handwritten Digit
Recognition, as well as a number of ways in which
neural networks were used for the recognition task?

 Main goal of this system?

 INTRODUCTION

 A handwritten digit recognition system was used to visualize

artificial neural networks. It is already widely used in the automatic
processing of bank cheques, postal addresses, in mobile phones etc

 To perform digit recognition, some basic knowledge on neural

network and image processing is needed. But, the customer may
use it without any prior knowledge in image processing or neural
network.

 Some of the existing systems include computational intelligence

techniques such as artificial neural networks or fuzzy logic, whereas
others may just be large lookup tables that contain possible
realizations of handwritten digits.
 Neural Networks

 Artificial neural networks have been developed since the 1940s.

 Artificial neural networks, usually called neural networks (NNs), are

systems composed of many simple processing elements (neurons)
operating in parallel whose function is determined by

1) Network Structure

2) Connection Strengths and

3) The Processing performed at Computing

elements or nodes.
 TWO MAIN GROUPS
OF
NEURAL NETWORK

SUPERVISED UN-SUPERVISED
LEARNING LEARNING
 Handwritten Digit
Recognition

Preprocessing:
It is used to eliminate extra problems due to non-class-
specific differences like

1) Size
2) Shear
3) Line thickness
4) Background and Digit colors
5) Resolution, etc.
 FEW TECHNIQUES TO
MINIMIZE THE DIFFERENCES

a) Deskewing:
The deskewing process works as follows:

1.An imaginary line is drawn through the digit:

This imaginary line can be described by an equation x = ay + b,
where x and y are the horizontal and vertical pixel index, respectively,
and a and b are line parameters.

2.A side slip is performed:

In formula: xnew = xold − ay for every pixel, resulting in the new line
x =b.

3.Slant is removed by performing rotation:

(a) left-leaning (b) side slip (c) rotation

(d) right-leaning (e) side slip (f) rotation

b) Thinning:
The digit images are thinned to a thickness of one pixel.

This can lead undesired side-effects ,such as spurs.

(a) threshold original (b) thinned (with spurs) (c) Pruned (by 5 pixels)
 The benefits of the thinning process far outweigh the negative
sides.

(d) smoothed (e) thinned

 The need for pruning can be avoided by Smoothing the

handwritten digit image before thinning, as seen in Figs. d and e.
c) Smoothing:

 Some acquisition methods may cause small

artifacts on the digit’s edges. These are
removed by a smoothing operation.

 Effective smoothing can be accomplished by

filling tiny gaps in both the foreground and
the background.
d) Resizing:

 Differences in digit size are due to different

image resolutions and handwriting styles.

 To minimize the effect of these differences, all

digits are resampled/rescaled to the same size in
pixels, before being fed to the recognizer.
 RESULTS
 A neural network, which is trained with 10000 handwritten digit
images from in image database, is shipped together with this
library as default neural network model.

 By performing benchmarking on 8920 handwritten digit images

from image database, the following recognition rate
is achieved.
RECOGNITION RATING TABLE
SET OF HANDWRITTEN DIGITS

a) TRAINING b) TESTING
Example for a single Digit
Recognition
Example for a String Recognition
 IPHONE-MOBILES
WITH HANDWRITTEN DIGIT
RECOGNITION FEATURE

BEFORE AFTER
 CONCLUSION

 The main goal is purely educational one, a moderate recognition

rate of 98% was reached on a test set.

 The Handwritten Digit Recognition is used to

recognize the Digits which are written by hand.

 The handwritten digit recognition have also proven to be a good

neural network architecture and application for the purpose of
introducing and demonstrating neural networks to the general
public.
 References

1) Kohonen, T.: Self-Organizing Maps. Springer-

Verlag, Berlin (1995)

2) Berend-Jan van der Zwaag Euregio -

Computational Intelligence Center - Dept. of
Electrical Engineering - University of Twente
Enschede, the Netherlands
 ANY
QUERIES ?
THANK
YOU