Optical character recognition without segmentation
M. Ozdil, и F. Vural. Document Analysis and Recognition, 1997., Proceedings of the Fourth
International Conference on, 2, стр. 483--486. IEEE Computer Society, (1997)
DOI: 10.1109/ICDAR.1997.620545
Аннотация
A segmentation-free approach for off-line optical character recognition
is presented. The proposed method performs the recognition by extracting
the characters from the whole word, avoiding the segmentation process.
A control point set which includes position and attribute vectors
is selected for the features. In the training mode, each sample character
is mapped to a set of control points and is stored in an archive
which belongs to an alphabet. In the recognition mode, the control
points of the input image are first extracted. Then, each control
point is matched to the control points in the alphabet according
to its attributes. During the matching process, a probability matrix
is constructed which holds some matching measures (probabilities)
for identifying the characters. Experimental results indicate that
the proposed method is very robust in extracting the characters from
a cursive script
%0 Conference Paper
%1 Ozdil1997
%A Ozdil, M.A.
%A Vural, F.T.Y.
%B Document Analysis and Recognition, 1997., Proceedings of the Fourth
International Conference on
%D 1997
%I IEEE Computer Society
%K (artificial algebra, archive, attribute character control cursive extraction extraction, feature group, image intelligence), learning matching matching, matrix matrix, measures, mode, off-line optical overlap point position probability probability, process, quality, recognition, script, segmentation-free set, training vectors, vectorsalphabet, word
%P 483--486
%R 10.1109/ICDAR.1997.620545
%T Optical character recognition without segmentation
%V 2
%X A segmentation-free approach for off-line optical character recognition
is presented. The proposed method performs the recognition by extracting
the characters from the whole word, avoiding the segmentation process.
A control point set which includes position and attribute vectors
is selected for the features. In the training mode, each sample character
is mapped to a set of control points and is stored in an archive
which belongs to an alphabet. In the recognition mode, the control
points of the input image are first extracted. Then, each control
point is matched to the control points in the alphabet according
to its attributes. During the matching process, a probability matrix
is constructed which holds some matching measures (probabilities)
for identifying the characters. Experimental results indicate that
the proposed method is very robust in extracting the characters from
a cursive script
@inproceedings{Ozdil1997,
abstract = {A segmentation-free approach for off-line optical character recognition
is presented. The proposed method performs the recognition by extracting
the characters from the whole word, avoiding the segmentation process.
A control point set which includes position and attribute vectors
is selected for the features. In the training mode, each sample character
is mapped to a set of control points and is stored in an archive
which belongs to an alphabet. In the recognition mode, the control
points of the input image are first extracted. Then, each control
point is matched to the control points in the alphabet according
to its attributes. During the matching process, a probability matrix
is constructed which holds some matching measures (probabilities)
for identifying the characters. Experimental results indicate that
the proposed method is very robust in extracting the characters from
a cursive script},
added-at = {2011-03-27T19:35:34.000+0200},
author = {Ozdil, M.A. and Vural, F.T.Y.},
biburl = {https://www.bibsonomy.org/bibtex/204a9ab012efec7f898d7389e3d081c1c/cocus},
booktitle = {Document Analysis and Recognition, 1997., Proceedings of the Fourth
International Conference on},
booktitleaddon = {Aug 18-20, 1997},
doi = {10.1109/ICDAR.1997.620545},
file = {:./00620545.pdf:PDF},
interhash = {db1e2ec0fdb79073abb75de8b2ea32c6},
intrahash = {04a9ab012efec7f898d7389e3d081c1c},
keywords = {(artificial algebra, archive, attribute character control cursive extraction extraction, feature group, image intelligence), learning matching matching, matrix matrix, measures, mode, off-line optical overlap point position probability probability, process, quality, recognition, script, segmentation-free set, training vectors, vectorsalphabet, word},
location = {#ieeeaddr#},
pages = {483--486},
publisher = {{IEEE} Computer Society},
timestamp = {2011-03-27T19:35:42.000+0200},
title = {Optical character recognition without segmentation},
volume = 2,
year = 1997
}