On-line recognition of handwritten characters
The On-line Recognition of Handwritten Characters
project in the Neural Networks Research Centre (NNRC) of
the Laboratory of
Computer and Information Science at Helsinki University of Technology is carried out in
co-operation with Nokia Research Center (NRC).
The members (former and current) of the research group are:
The goal of the project is to develop adaptive methods for on-line recognition
of handwritten characters. In this case, adaptation means that the
system is able to learn new writing styles during its normal
use. Due to the learning, the user can use his own natural style
of writing instead of some constrained style. Typical
applications for on-line handwriting recognition are small
hand-held devices, for example PDAs, in which it replaces a
keyboard. Our work concentrates on the recognition of isolated
alphanumeric characters.
The recognition and adaptation methods applied so far are described in
the following publications and papers:
Matti Aksela and Jorma Laaksonen,
Using diversity of errors for selecting members of a committee
classifier.
Pattern Recognition,
Vol. 39, Number 4, pp. 608-623, 2006.
Matti Aksela and Jorma Laaksonen,
On Adaptive Confidences for
Critic-Driven Classifier Combining,
in Proceedings of ICAPR 2005, pp. 71-80, 2005.
Matti Aksela.
Comparison of Classifier Selection Methods for Improving
Committee Performance, in Proceedings of MCS2003, pp. 84-93, 2003.
Matti Aksela, Ramunas Girdziusas, Jorma Laaksonen, Erkki Oja,
Jari Kangas. Methods for adaptive combination of classifiers with
application to recognition of handwritten characters.
International Journal on Document Analysis and Recognition,
Vol. 6 Number 1 (August 2003) pp.23-41, 2003.
Vuokko Vuori. Adaptive
Methods for On-Line Recognition of Isolated Handwritten
Characters . Doctoral Thesis, Helsinki University of
Technology, 2002.
Matti Aksela, Ramunas Girdziusas, Jorma Laaksonen, Erkki
Oja. Class-Confidence Critic
Combining. Published in the Proceedings of the 8th
International Workshop on Frontiers in Handwriting
Recognition, pp. 201-206, Niagara-on-the Lake, Canada, August
2002.
Vuokko Vuori. Clustering Writing
Styles with a Self-Organizing
Map. Published in the Proceedings of the 8th International
Workshop on Frontiers in Handwriting Recognition, pp. 345-350,
Niagara-on-the Lake, Canada, August 2002.
Vuokko Vuori, Jorma Laaksonen. A
Comparison of Techniques for
Automatic Clustering of Handwritten Characters. Published in
the Proceedings of the 16th International Conference on
Pattern Recognition, volume 3, pp. 168-171, Quebec, Canada,
August 2002.
Matti Aksela, Jorma Laaksonen, Erkki Oja, Jari Kangas.
Application of adaptive committee classifiers in on-line character recognition.
Published in the Proceedings of ICAPR'01, pp 270-279, 2001.
Matti Aksela, Jorma Laaksonen, Erkki Oja, Jari Kangas.
Rejection methods for an adaptive committee classifier. Published in the Proceedings of ICDAR'01, pp 982-986, Seattle, USA, September 2001.
Vuokko Vuori, Jorma Laaksonen, Erkki Oja, Jari Kangas.
Speeding up On-line Recognition of Handwritten Characters by Pruning the Prototype Set. Published in the Proceedings of ICDAR'01, pp 501-505, Seattle, USA, September 2001.
Vuokko Vuori, Jorma Laaksonen, Jari Kangas.
Influence of Erroneous Learning Samples on Adaptation in On-line Handwriting Recognition. Published in Pattern Recognition, pp 915-926, volume 35, number 4, 2002.
Vuokko Vuori, Jorma Laaksonen, Erkki Oja, Jari Kangas. Experiments
with adaptation strategies for a prototype-based recognition system
for isolated handwritten characters. Published in
International Journal on Document Analysis and Recognition,
pp 150-159, volume 3, issue 2, 2001.
Vuokko Vuori, Erkki Oja. Analysis of Different
Writing Styles with the Self-Organizing
Map.
Published in the proceedings of the ICONIP2000, volume 2, pp 1243-1247, Taejon, South Korea,
November 2000.
Vuokko Vuori, Jorma Laaksonen, Erkki Oja, Jari Kangas.
Controlling On-Line Adaptation of a Prototype-Based
Classifier for Handwritten Characters. Published in the
proceedings of the ICPR2000, volume 2, pp 331-334, Barcelona, Spain, September 2000.
Vuokko Vuori, Matti Aksela, Jorma Laaksonen, Erkki Oja, Jari Kangas.
Adaptive character recognizer for a hand-held
device: implementation and evaluation
setup
Published in the proceedings of the IWFHR2000, pp 13-22, Amsterdam, The
Netherlands, September 2000.
Matti Aksela.
Handwritten Character Recognition: A Palm-top Implementation and
Adaptive Committee Experiments. Master's Thesis, Helsinki University of Technology, 2000.
Jorma Laaksonen, Vuokko Vuori, Matti Aksela, Erkki Oja, Jari Kangas.
Experiments with Adaptation Methods in On-line
Recognition of Isolated Latin Characters. To appear in book
Cognitive and Neural Models for Word Recognition and Document
Processing. World Scientific Press.
Vuokko Vuori, Jorma Laaksonen, Erkki Oja, Jari Kangas. On-line Adaptation in Recognition of Handwritten
Alphanumeric Characters. Published in Proceedings of ICDAR
'99, pp 792-795, Bangalore, India, September 1999.
Jorma Laaksonen, Matti Aksela, Erkki Oja, Jari Kangas. Dynamically
Expanding Context as Committee Adaptation Method in On-Line
Recognition of Handwritten Latin Characters. Published in
Proceedings of ICDAR '99, pp 796-799, Bangalore, India, September 1999.
Jorma Laaksonen, Matti Aksela, Erkki Oja, Jari Kangas. Adaptive Local
Subspace Classifier in On-Line Recognition of Handwritten
Characters. Published in Proceedings of IJCNN'99, volume 4, pp 2812-2815, 1999.
Jorma Laaksonen, Vuokko Vuori, Erkki Oja, Jari Kangas. Adaptation of
Prototype Sets in On-line Recognition of Isolated Handwritten Latin
Characters. Published in book Advances in Handwriting
Recognition, World Scientific Publishing Co., pp 489-497, 1999.
Vuokko Vuori. Adaptation in on-line recognition of
handwriting. Master's Thesis, Helsinki University of Technology,
January 1999.
Jorma Laaksonen, Jarmo Hurri, Erkki Oja, Jari Kangas. Comparison of
Adaptive Strategies for On-Line Character Recognition. Published
in Proceedings of ICANN'98, pp 245-250, Skövde, Sweden, September 1998.
Jorma Laaksonen, Jarmo Hurri, Erkki Oja, Jari Kangas. Experiments
with a Self-Supervised Adaptive Classification Strategy in On-Line
Recognition of Isolated Handwritten Latin Characters. Published in
Proceedings of IWFHR98, pp 475-484, Taejon, South Korea, August 1998.
For further information, please contact
vuokko.vuori@hut.fi .
Last modified 2.9.2002.