AUTOMATIC RECOGNITION OF
HANDWRITTEN IMAGE METHOD BASED ON RECURRENT NEURAL NETWORK
abstract
The
automatic recognition of handwritten text - such as letters, manuscripts or
entire books - has been a focus of intensive research for several decades. Yet
the Problem is far
from being solved.
Particularly
in the field of unconstrained Handwriting recognition where the writing styles of various writers must
be dealt
With,
severe difficulties are encountered.
Making handwritten
texts available for Searching
and browsing is of tremendous value. For example, one might be interested in finding all
occurrences of the word "complain" in the letters sent to a Company.
As another example, libraries all over the
world store huge numbers of handwritten books that are of crucial importance for preserving the
world's cultural
Heritage.
Making these books
available for searching and browsing would greatly help researchers and the
public alike. Certain efforts have already been put into Word spotting for
historical data.
Another related
application is the segmentation of Images of historical documents into
meaningful regions, which can be improved with keyword spotting.
In the keyword "Fig." is spotted in
the images to help
Identifying
figures and their corresponding captions.
Finally, it is
worth mentioning That
Google and Yahoo have announced their intention to make handwritten books
accessible through their search engines.
In this context, keyword spotting will be a Valuable
tool for users browsing the contents of these books.
Transcribing the Entire text of a hand written
document for searching is not only inefficient as far as Computational costs are
concerned, but it may also result in poor performance,
Since
mix-recognized words cannot be found.
Therefore, techniques especially
Designed for the task of keyword spotting have been developed. Next, were
views? Related work
from this area.
No comments:
Post a Comment