Optical Character Recognition for Cursive Handwriting
Nafiz Arica, Student Member, IEEE, and Fatos T. Yarman-Vural, Senior Member, IEEE AbstractÐIn this paper, a new analytic scheme, which uses a sequence of segmentation and recognition algorithms, is proposed for offline cursive handwriting recognition problem. First, some global parameters, such as slant angle, baselines, and stroke width and height are estimated. Second, a segmentation method finds character segmentation paths by combining gray scale and binary information. Third, Hidden Markov Model (HMM) is employed for shape recognition to label and rank the character candidates. For this purpose, a string of codes is extracted from each segment to represent the character candidates. The estimation of feature space parameters is embedded in HMM training stage together with the estimation of the HMM model parameters. Finally, the lexicon information and HMM ranks are combined in a graph optimization problem for word-level recognition. This method corrects most of the errors produced by segmentation and HMM ranking stages by maximizing an information measure in an efficient graph search algorithm. The experiments in dicate higher recognition rates compared to the available methods reported in the literature. Index TermsÐHandwritten word recognition, preprocessing, segmentation, optical character recognition, cursive handwriting, hidden Markov model, search, graph, lexicon matching.
HE most difficult problem in the field of Optical Character Recognition (OCR) is the recognition of unconstrained cursive handwriting. The present tools for modeling almost infinitely many variations of human handwriting are not yet sufficient. The similarities of distinct character shapes, the overlaps, and interconnection of the neighboring characters further complicate the problem. Additionally, when observed in isolation, characters are often ambiguous and require context information to reduce the classification error. Thus, current research aims at developing constrained systems for limited domain applications such as postal address reading , check sorting , tax reading , and office automation for text entry . A well-defined lexicon plus a well-constrained syntax help provide a feasible solution to the problem . Handwritten Word Recognition techniques use either holistic or analytic strategies for training and recognition stages. Holistic strategies employ top-down approaches for recognizing the whole word, thus eliminating the segmentation problem . In this strategy, global features, extracted from the entire word image, are used in recognition of limited-size lexicon. As the size of the lexicon gets larger, the complexity of algorithms increase linearly due to the need for a larger search space and a more complex pattern representation. Additionally, the recognition rates decrease rapidly due to the decrease in betweenclass-variances in the feature space. The analytic strategies, on the other hand, employ bottom-up approaches, starting from stroke or character-
level and going towards producing a meaningful text. Explicit  or implicit  segmentation of word into characters or strokes is required for this strategy. With the cooperation of segmentation stage, the problem is reduced to the recognition of simple isolated characters or strokes, which can be handled for unlimited vocabulary. However, there is no segmentation algorithm available in the literature for correctly extracting the characters from a given word image. The popular techniques are based on over-segmenting the words and applying a search algorithm for grouping segments to make up characters , . If a lexicon of limited size is given, dynamic programming is used to rank every word in the lexicon. The word with the highest rank is chosen as the recognition hypothesis. The complexity of search process for this strategy also increases linearly with the...
References:  N. Arica and F.T. Yarman-Vural, ªOne Dimensional Representation Of Two Dimensional Information For HMM Based Handwritten Recognition,º Pattern Recognition Letters, vol. 21, pp. 583592, 2000. N. Arica and F.T. Yarman-Vural, ªA New Scheme for Off-Line Handwritten Connected Digit Recognition,º Proc. Int 'l Conf. Pattern Recognition, pp. 1127-1131, 1998. A. Atici and F.T. Yarman-Vural, ªA Heuristic Method for Arabic Character Recognition,º J. Signal Processing, vol. 62, pp. 87-99, 1997. R.G. Casey and E. Lecolinet, ªStrategies in Character Segmentation: A Survey,º Proc. Third Int 'l Conf. Document Analysis and Recognition, pp. 1028-1033, 1995. T. Caesar, J.M. Gloger, and E. Mandler, ªEstimating The Baseline For Written Material,º Proc. Third Int 'l Conf. Document Analysis and Recognition, pp. 382-385, 1995. A. Dengel, R. Hoch, F. Hones, T. Jager, M. Malburg, and A. Weigel, ªTechniques For Improving OCR Results,º Handbook of Character Recognition and Document Image Analysis, H. Bunke and P.S.P. Wang, eds., pp. 227-254, 1997. S. Gopisetty, R. Lorie, J. Mao, M. Mohiuddin, A. Sorin, and E. Yair, ªAutomated Forms-Processing Software and Services,º IBM J. Research and Development, vol. 40, no. 2, pp. 211-230, 1996. N. Gorski, V. Anisimov, E. Augustin, O. Baret, D. Price, and J.-C. Simon, ªA2iA Check Reader: A Family of Bank Check Recognition Systems,º Proc. Fifth Int 'l Conf. Document Analysis and Recognition, pp. 523-526, 1999. D. Guillevic and C.Y. Suen, ªRecognition of Legal Amounts on Bank Cheques,º Pattern Analysis and Applications, vol. 1, no. 1, pp. 28-41, 1998. G. Kim and V. Govindaraju, ªA Lexicon Driven Approach to Handwritten Word Recognition for Real-Time Applications,º IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 19, no. 4, pp. 366-379, 1997. G. Kim, V. Govindaraju, and S.N. Srihari, ªAn Architecture For Handwritten Text Recognition Systems,º Int 'l J. Document Analysis and Recognition, vol. 2, no. 1, pp. 37-44, 1999. A. Kornai, K.M. Mohiuddin, and S.D. Connell, ªRecognition of Cursive Writing on Personal Checks,º Proc. Int 'l Workshop Frontiers in Handwriting Recognition, pp. 373-378, 1996. W. Lee, D.J. Lee, and H.S. Park, ªA New Methodology for Gray Scale Character Segmentation and Recognition,º IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 18, no. 10, pp. 10451050, Oct. 1996.
 J. Mao, P. Sinha, and K. Mohiuddin, ªA System For Cursive Handwritten Address Recognition,º Proc. Int 'l Conf. Pattern Recognition, pp. 1285-1287, 1998.  S. Madhvanath, G. Kim, and V. Govindaraju, ªChaincode Contour Processing for Handwritten Word Recognition,º IEEE Trans. Pattern Recognition and Machine Intelligence, vol. 21, no. 9, pp. 928932, Sept. 1999.  M. Mohamed and P. Gader, ªHandwritten Word Recognition Using Segmentation-Free Hidden Markov Modeling and Segmentation Based Dynamic Programming Techniques,º IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 18, no. 5, pp. 548-554, May 1996.  L.R. Rabiner, ªA Tutorial on Hidden Markov Models and Selected Applications in Speech Recognition,º Proc. IEEE, vol. 77, pp. 257286, 1989.  A.W. Senior and A.J. Robinson, ªAn Off-Line Cursive Handwriting Recognition System,º IEEE Trans. Pattern Recognition and Machine Intelligence, vol. 20, no. 3, pp. 309-322, 1998.  M. Shridhar, G. Houle, and F. Kimura, ªHandwritten Word Recognition Using Lexicon Free and Lexicon Directed Word Recognition Algorithms,º Proc. Fourth Int 'l Conf. Document Analysis and Recognition, pp. 861-866, 1997.  S.N. Srihari, Y.C. Shin, Y. Ramanaprasad, and D.S. Lee, ªA System To Read Names and Adresses on Tax Forms,º Proc. IEEE, vol. 84, no. 7, pp. 1038-1049, 1996.  S.N. Srihari and E.J. Keubert, ªIntegration of Handwritten Address Interpretation Technology into the United states Postal service Remote Computer Reader System,º Proc. Fourth Int 'l Conf. Document Analysis and Recognition, pp. 892-896, 1997.  é.D. Trier and A.K. Jain, ªGoal Directed Evaluation of Binarization Methods,º IEEE Trans. Pattern Recognition and Machine Intelligence, vol. 17, no. 12, pp. 1191-1201, 1995.  J. Wang and J. Jean, ªSegmentation of Merged Characters by Neural Networks and Shortest Path,º Pattern Recognition, vol. 27, no. 5, pp. 649, 1994. Nafiz Arica received the BSc degree from the Turkish Naval Academy in 1991. He worked for the Navy as communications and combat officer for four years. In 1995, he joined theMiddle East Technical University (METU) where he received the MSc degree in computer engineering. His thesis was awarded the thesis of the year in 1998 at METU. He is currently a PhD candidate in the Computer Engineering Department and also a lieutenant in the Navy. His research interests include character recognition, content-based image representation. He is a student member of the IEEE. Fatos T. Yarman-Vural received the BSc degree with honors in electrical engineering from the Technical University of Istanbul in 1973, the MSc degree in electrical engineering from Bogazici University in 1975, and the PhD degree in electrical engineering and computer science from Princeton University in 1981. From 1981 to 1983, she was a research scientist in Marmara Research Institute in Turkey. From 1983 to 1985, she was a visiting professor at Drexel University. From 1985 to 1992, she was a technical and deputy manager at Yapitel Inc. Since 1992, she has been a professor in the Computer Engineering Department of Middle East Technical University (METU). Her research area covers computer vision, image processing, pattern recognition, and artificial intelligence. She has been involved in teaching, consulting, and organizing conferences in these areas. She was the chair woman in the Computer Engineering Department between 1996-2000. Currently, she is assistant to the president at METU. She is a senior member of IEEE and a member of Turkish Informatics Foundation, Turkish Information Association, and chamber of Electrical Engineers in Turkey.
    
  
Please join StudyMode to read the full document