Advanced

Core Points - Variable and Reduced Parameterization for Symbol Recognition

Sternby, Jakob LU (2005)
Abstract
Recent research in the field of on-line handwriting recognition has

been focused on statistical systems such as Hidden Markov Models,

Neural Networks or a combination of these. There are however merits

of employing an approach based on template matching. The first part

of this thesis presents a new strategy for parameterization of

on-line handwritten character samples. A novel efficient template

matching method enabled by this parameterization is also proposed.

In consecutive chapters of the thesis it is also shown that the

proposed structural parameterization enables an effective

application of template matching methods to the recognition of

cursive... (More)
Recent research in the field of on-line handwriting recognition has

been focused on statistical systems such as Hidden Markov Models,

Neural Networks or a combination of these. There are however merits

of employing an approach based on template matching. The first part

of this thesis presents a new strategy for parameterization of

on-line handwritten character samples. A novel efficient template

matching method enabled by this parameterization is also proposed.

In consecutive chapters of the thesis it is also shown that the

proposed structural parameterization enables an effective

application of template matching methods to the recognition of

cursive script. Ambiguity of the shapes of individual characters in

unconstrained cursive handwriting necessitates dictionary

interaction for real applications. A fast technique for applying

dictionary information to the language independent graph approach

has also been developed. A large data set of on-line cursive writing

has been collected and the developed system for mixed and cursive

on-line handwriting recognition has been shown to produce state of

the art results on this data set. One of the obvious potential

weaknesses of a structural parameterization technique such as the

one presented in this thesis is its sensitivity to digital noise in

the form of superfluous coordinates. Possible remedies to deal with

such effects have also been studied. (Less)
Please use this url to cite or link to this publication:
author
supervisor
organization
publishing date
type
Thesis
publication status
published
subject
pages
77 pages
ISBN
91-628-6580-3
language
English
LU publication?
yes
id
9e3b52c7-721b-4231-9fbf-e450881309d2 (old id 635695)
date added to LUP
2009-05-25 16:06:36
date last changed
2016-09-19 08:45:00
@misc{9e3b52c7-721b-4231-9fbf-e450881309d2,
  abstract     = {Recent research in the field of on-line handwriting recognition has<br/><br>
been focused on statistical systems such as Hidden Markov Models,<br/><br>
Neural Networks or a combination of these. There are however merits<br/><br>
of employing an approach based on template matching. The first part<br/><br>
of this thesis presents a new strategy for parameterization of<br/><br>
on-line handwritten character samples. A novel efficient template<br/><br>
matching method enabled by this parameterization is also proposed.<br/><br>
In consecutive chapters of the thesis it is also shown that the<br/><br>
proposed structural parameterization enables an effective<br/><br>
application of template matching methods to the recognition of<br/><br>
cursive script. Ambiguity of the shapes of individual characters in<br/><br>
unconstrained cursive handwriting necessitates dictionary<br/><br>
interaction for real applications. A fast technique for applying<br/><br>
dictionary information to the language independent graph approach<br/><br>
has also been developed. A large data set of on-line cursive writing<br/><br>
has been collected and the developed system for mixed and cursive<br/><br>
on-line handwriting recognition has been shown to produce state of<br/><br>
the art results on this data set. One of the obvious potential<br/><br>
weaknesses of a structural parameterization technique such as the<br/><br>
one presented in this thesis is its sensitivity to digital noise in<br/><br>
the form of superfluous coordinates. Possible remedies to deal with<br/><br>
such effects have also been studied.},
  author       = {Sternby, Jakob},
  isbn         = {91-628-6580-3},
  language     = {eng},
  pages        = {77},
  title        = {Core Points - Variable and Reduced Parameterization for Symbol Recognition},
  year         = {2005},
}