Abstract
In this paper several aspects of a recognition system for cursive handwritten German address words (cities and streets) are described. The recognition system is based on Hidden Markov Models (HMMs), whereat the focus is on two main problems: the changes in the writing style depending on time or regional differences and the difficulty to select the correct (complete) dictionary for address reading. The first problem leads to the examination of three different adaptation techniques: Maximum Likelihood (ML), Maximum Likelihood Linear Regression (MLLR) and Scaled Likelihood Linear Regression (SLLR). To handle the second problem language models based on backoff character n-grams are examined to evaluate the performance of an open vocabulary recognition (without dictionary). For both problems the determination of confidence measures (based on the frame-normalized likelihood, a garbage model, a two-best recognition or an unconstrained character decoding) is important, either for an unsupervised adaptation or the detection of out of vocabulary words (OOV). The databases, which are used for recognition, are provided by Siemens Dematic (SD) within the Adaptive READ project.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Bazzi, I., Schwartz, R., Makhoul, J.: An Omnifont Open-Vocabulary OCR System for English and Arabic. IEEE Transactions on Pattern Analysis and Machine Intelligence 21(6), 495–504 (1999)
Brakensiek, A., Kosmala, A., Rigoll, G.: Comparing Adaptation Techniques for On-Line Handwriting Recognition. In: 6th Int. Conference on Document Analysis and Recognition (ICDAR), Seattle, USA, September 2001, pp. 486–490 (2001)
Brakensiek, A., Rigoll, G.: A Comparison of Character N-Grams and Dictionaries Used for Script Recognition. In: 6th Int. Conference on Document Analysis and Recognition (ICDAR), Seattle, USA, September 2001, pp. 241–245 (2001)
Brakensiek, A., Rottland, J., Rigoll, G.: Handwritten Address Recognition with OpenVocabulary Using Charcter N-Grams. In: 8th Int. Workshop on Frontiers in Handwriting Recognition (IWFHR), Niagara-on-the-Lake, Canada, August 2002, pp. 357–362 (2002)
Brakensiek, A., Rottland, J., Rigoll, G.: Confidence Measures for an Address Reading System. In: 7th Int. Conference on Document Analysis and Recognition (ICDAR), Edinburgh (August 2003)
Brakensiek, A., Rottland, J., Wallhoff, F., Rigoll, G.: Adaptation of an Address Reading System to Local Mail Streams. In: 6th Int. Conference on Document Analysis and Recognition (ICDAR), Seattle, USA, September 2001, pp. 872–876 (2001)
Brakensiek, A., Willett, D., Rigoll, G.: Unlimited Vocabulary Script Recognition Using Character N-Grams. In: 22. DAGM-Symposium, Tagungsband, Kiel, Germany, September 2000, pp. 436–443. Springer, Heidelberg (2000)
Caesar, T., Gloger, J.M., Mandler, E.: Preprocessing and Feature Extraction for a Handwriting Recognition System. In: Proc. Int. Conference on Document Analysis and Recognition (ICDAR), Tsukuba, Japan, October 1993, pp. 408–411 (1993)
Clarkson, P., Rosenfeld, R.: Statistical Language Modeling Using the CMU-Cambridge Toolkit. In: 5th European Conference on Speech Communication and Technology (Eurospeech), Rhodes, Greece, September 1997, pp. 2707–2710 (1997)
Dolfing, J.G.A., Wendemuth, A.: Combination of Confidence Measures in Isolated Word Recognition. In: 5th Int. Conference on Spoken Language Processsing (ICSLP), Sydney, Australia, December 1998, pp. 3237–3240 (1998)
Franke, J., Gloger, J.M., Kaltenmeier, A., Mandler, E.: A Comparison of Gaussian Distribution and Polynomial Classifiers in a Hidden Markov Model Based System for the Recognition of Cursive Script. In: Proc. Int. Conference on Document Analysis and Recognition (ICDAR), Ulm, Germany, August 1997, pp. 515–518 (1997)
Gauvain, J.-L., Lee, C.-H.: Maximum a Posteriori Estimation for Multivariate Gaussian Mixture Observation of Markov Chains. IEEE Transactions on Speech and Audio Processing 2(2), 291–298 (1994)
Gloger, J.M., Kaltenmaier, A., Mandler, E., Andrews, L.: Reject Management in a Handwriting Recognition System. In: Int. Conference on Document Analysis and Recognition (ICDAR), Ulm, Germany, August 1997, pp. 556–559 (1997)
Hazen, T.J., Bazzi, I.: A Comparison and Combination of Methods for OOV Word Detection and Word Confidence Scoring. In: IEEE Int. Conference on Acoustics, Speech, and Signal Processing (ICASSP), Salt Lake City, Utah (May 2001)
Jelinek, F.: Statistical Methods for Speech Recognition. The MIT Press, Cambridge (1998)
Katz, S.M.: Estimation of probabilities from sparse data for the language model component of a speech recognizer. IEEE Transactions on Acoustic, Speech and Signal Processing 35(3), 400–401 (1987)
Leggetter, C.J., Woodland, P.C.: Speaker Adaptation of Continuous Density HMMs using Multivariate Linear Regression. In: Int. Conference on Spoken Language Processing (ICSLP), Yokohama, Japan, September 1994, pp. 451–454 (1994)
Lu, Z., Bazzi, I., Kornai, A., Makhoul, J., Natarajan, P., Schwartz, R.: A Robust, Language- Independent OCR System. In: Proc. 27th AIPR Workshop: Advances in Computer-Assisted Recognition (SPIE), pp. 96–105 (1999)
Mandler, E., Oberländer, M.: A single pass algorithm for fast contour coding of binary images. In: 12.DAGM-Symposium, Tagungsband, Oberkochen- Aalen, Germany, September 1990, pp. 248–255. Springer, Heidelberg (1990)
Marti, U.-V., Bunke, H.: Unconstrained Handwriting Recognition: Language Models, Perplexity, and System Performance. In: 7th Int. Workshop on Frontiers in Handwriting Recognition (IWFHR), Amsterdam, Netherlands, September 2000, pp. 463–468 (2000)
Marukatat, S., Artieres, T., Gallinari, P.: Rejection measures for Handwriting sentence Recognition. In: 8th Int. Workshop on Frontiers in Handwriting Recognition (IWFHR), Niagara-on-the-Lake, Canada, August 2002, pp. 24–29 (2002)
Miletzki, U., Bayer, T., Schäfer, H.: Continuous Learning Systems: Postal Address Readers with built-in learning capability. In: 5th Int. Conference on Document Analysis and Recognition (ICDAR), Bangalore, India, pp. 329–332 (1999)
Paul, D.B.: An efficient A* stack decoder algorithm for continuous speech recognition with a stochastic language model. In: IEEE Int. Conference on Acoustics, Speech, and Signal Processing (ICASSP), San Francisco, CA, March 1992, pp. 25–28 (1992)
Pitrelli, J., Perrone, M.: Confidence Modeling for Verification Post-Processing for Handwriting Recognition. In: 8th Int. Workshop on Frontiers in Handwriting Recognition (IWFHR), Niagara-on-the-Lake, Canada, August 2002, pp. 30–35 (2002)
Pitrelli, J.F., Ratzlaff, E.H.: Quantifying the Contribution of Language Modeling to Writer- Independent On-Line Handwriting Recognition. In: 7th Int. Workshop on Frontiers in Handwriting Recognition (IWFHR), Amsterdam, Netherlands, September 2000, pp. 383–392 (2000)
Rabiner, L.R., Juang, B.H.: An Introduction to Hidden Markov Models. IEEE ASSP Magazine, 4–16 (1986)
Rose, R., Paul, D.: A Hidden Markov Model based Keyword Recognition System. In: IEEE Int. Conference on Acoustics, Speech, and Signal Processing (ICASSP), Albuquerque, New Mexico, pp. 129–132 (1990)
Rosenfeld, R.: Two decades of statistical language modeling: Where do we go from here? Proceedings of the IEEEÂ 88(8) (2000)
Schukat-Talamazzini, E.G.: Automatische Spracherkennung - Grundlagen, statistische Modelle und effiziente Algorithmen,s Vieweg, Braunschweig (1995)
Schüßler, M., Niemann, H.: A HMM-based System for Recognition of Handwritten Adress Words. In: 6th Int. Workshop on Frontiers in Handwriting Recognition (IWFHR), Taejon, Korea, pp. 505–514 (1998)
Senior, A., Nathan, K.: Writer adaptation of a HMM handwriting recognition system. In: IEEE Int. Conference on Acoustics, Speech, and Signal Processing (ICASSP), Munich, Germany, April 1997, pp. 1447–1450 (1997)
Shridhar, M., Kimura, F., Truijen, B., Houle, G.F.: Impact of Lexicon Completeness on City Name Recognition. In: 8th Int. Workshop on Frontiers in Handwriting Recognition (IWFHR), Niagara-on-the-Lake, Canada, August 2002, pp. 513–518 (2002)
Steinherz, T., Rivlin, E., Intrator, N.: Offline cursive sript word recognition - a survey. Int. Journal on Document Analysis and Recognition (IJDAR) 2, 90–110 (1999)
Tomai, C.I., Allen, K.M., Srihari, S.N.: Recognition of Handwritten Foreign Mail. In: 6th Int. Conference on Document Analysis and Recognition (ICDAR), Seattle, USA, September 2001, pp. 882–886 (2001)
Valtchev, V., Odell, J.J., Woodland, P.C., Young, S.J.: Lattice-Based Discriminative Training for Large Vocabulary Speech Recognition Systems. In: IEEE Int. Conference on Acoustics, Speech, and Signal Processing (ICASSP), Atlanta, GA, May 1996, pp. 605–608 (1996)
Vinciarelli, A., Bengio, S.: Writer adaptation techniques in HMM based Off-Line Cursive Script Recognition. Pattern Recognition Letters 23(8), 905–916 (2002)
Wallhoff, F., Willett, D., Rigoll, G.: Frame Discriminative and Confidence- Driven Adaptation for LVCSR. In: IEEE Int. Conference on Acoustics, Speech, and Signal Processing (ICASSP), Istanbul, Turkey, June 2000, pp. 1835–1838 (2000)
Willett, D., Neukirchen, C., Rigoll, G.: DUCODER-The Duisburg University LVSCR Stackdecoder. In: Proc. IEEE Int. Conf. on Acoustics, Speech, and Signal Processing (ICASSP) (June 2000)
Williams, G., Renals, S.: Confidence measures from local posterior probability estimates. Computer Speech and Language 13, 395–411 (1999)
Young, S.: Detecting Misrecognitions and Out-Of Vocabulary Words. In: IEEE Int. Conference on Acoustics, Speech, and Signal Processing (ICASSP), Adelaide, Australia, April 1994, pp. 21–24 (1994)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2004 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
Brakensiek, A., Rigoll, G. (2004). Handwritten Address Recognition Using Hidden Markov Models. In: Dengel, A., Junker, M., Weisbecker, A. (eds) Reading and Learning. Lecture Notes in Computer Science, vol 2956. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-24642-8_7
Download citation
DOI: https://doi.org/10.1007/978-3-540-24642-8_7
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-21904-0
Online ISBN: 978-3-540-24642-8
eBook Packages: Springer Book Archive