A Neural Approach to Cursive Handwritten Character Recognition Using Features Extracted from Binarization Technique - Complex System Modelling and Control Through Intelligent Soft Computations - page 731

1 Introduction

In today s life, people are constantly writing notes on paper. These notes range from

the minutes of a meeting, a short reminder, a list of things to do, to a long-winded
letter to a friend. In this digital age, some may wonder why people still use paper.
The answer is simple: an effective substitute for paper has not been invented. Paper
is still

'

ciency, affordability, usability, and
mobility. Yet having a digital copy of these notes has many benefits, most notably
the ability to search and organize the notes in many ways instantaneously. Being
able to search a huge database of written notes saves time and money which are the
two important things in the business world. People need a way to convert hand-
written text to digital text, which can be searched and organized however the user
wants. A common complaint and excuse of people is that they couldn

the champion when it comes to ef

t read their
own handwriting. That makes us ask ourselves the question: If people sometimes
can

'

t read their own handwriting, with which they are quite familiar, what chance
does a computer have? Fortunately, there are powerful tools that can be used that
are easily implementable on a computer.
Character recognition is the ability of a computer to receive and interpret
handwritten input from sources such as paper documents, photographs, touch-
panels, light pen and other devices. This technology is steadily growing toward its
maturity. The domain of hand written text recognition has two completely different
problems of On-line and Off-line character recognition. On-line character recog-
nition (Bharath and Madhvanath 2008 ) involves the automatic conversion of
characters as it is written on a special digitizer or PDA, where a sensor picks up the
pen-tip movements as well as pen-up/pen-down switching. That kind of data is
known as digital ink and can be regarded as a dynamic representation of hand-
written characters. The obtained signal is converted into letter codes which are
usable within computer and text-processing applications. On the contrary, off-line
character recognition involves the automatic conversion of character (as an image)
into letter codes which are usable within computer and text-processing applications.
The data obtained by this form is regarded as a static representation of handwritten
character. The technology is successfully used by businesses which process lots of
handwritten documents, like insurance companies. The quality of recognition
can be substantially increased by structuring the document (by using forms). The
off-line character recognition is comparatively dif

'

cult, as different people have
different handwriting styles and also the characters are extracted from documents of
different intensity and background (Farooq et al. 2008 ). Limiting the range of input
can allow recognition process to improve.
Feed Forward Neural Network plays a great role in Medical Diagnostics (Azar
2013 ; Azar and El-Said 2013 ). The most important type of feed forward neural
network is the Back Propagation Neural Network (BPNN). Back Propagation is a
systematic method for training multi-layer arti

cial neural network (Sivanandam
and Deepa 2008 ). It is a multilayer feed forward network using gradient descent