amazon

Saturday, August 22, 2015

AUTOMATIC RECOGNITION OF HANDWRITTEN IMAGE METHOD BASED ON RECURRENT NEURAL NETWORK

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                             Another related application is the segmentation of Images of historical documents into meaningful regions, which can be improved with keyword spotting.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