A SVM-based cursive character recognizer

This paper presents a cursive character recognizer, a crucial module in any cursive word recognition system based on a segmentation and recognition approach. The character classification is achieved by using support vector machines (SVMs) and a neural gas. The neural gas is used to verify whether lower and upper case version of a certain letter can be joined in a single class or not. Once this is done for every letter, the character recognition is performed by SVMs. A database of 57293 characters was used to train and test the cursive character recognizer. SVMs compare notably better, in terms of recognition rates, with popular neural classifiers, such as learning vector quantization and multi-layer-perceptron. SVM recognition rate is among the highest presented in the literature for cursive character recognition.

[1]  Francesco Camastra,et al.  Cursive character recognition by learning vector quantization , 2001, Pattern Recognit. Lett..

[2]  Jonathan J. Hull,et al.  A Database for Handwritten Text Recognition Research , 1994, IEEE Trans. Pattern Anal. Mach. Intell..

[3]  Ralf Herbrich,et al.  Learning Kernel Classifiers: Theory and Algorithms , 2001 .

[4]  Jorma Laaksonen,et al.  LVQ_PAK: The Learning Vector Quantization Program Package , 1996 .

[5]  Sargur N. Srihari,et al.  On-Line and Off-Line Handwriting Recognition: A Comprehensive Survey , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[6]  Michael A. Arbib,et al.  The handbook of brain theory and neural networks , 1995, A Bradford book.

[7]  Ehud Rivlin,et al.  Offline cursive script word recognition – a survey , 1999, International Journal on Document Analysis and Recognition.

[8]  Michael Blumenstein,et al.  Experimental analysis of the modified direction feature for cursive character recognition , 2004, Ninth International Workshop on Frontiers in Handwriting Recognition.

[9]  Thomas Martinetz,et al.  Topology representing networks , 1994, Neural Networks.

[10]  Alexander J. Smola,et al.  Learning with kernels , 1998 .

[11]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[12]  Shimon Ullman,et al.  Reading cursive handwriting by alignment of letter prototypes , 1991, International Journal of Computer Vision.

[13]  Y. Miyake,et al.  Machine and human recognition of segmented characters from handwritten words , 1997, Proceedings of the Fourth International Conference on Document Analysis and Recognition.

[14]  Nello Cristianini,et al.  Kernel Methods for Pattern Analysis: List of kernels , 2004 .

[15]  Sargur N. Srihari,et al.  Off-Line Cursive Script Word Recognition , 1989, IEEE Trans. Pattern Anal. Mach. Intell..

[16]  C. Scagliola,et al.  Generalised projections: a tool for cursive handwriting normalisation , 1999, Proceedings of the Fifth International Conference on Document Analysis and Recognition. ICDAR '99 (Cat. No.PR00318).

[17]  M. Stone.  Cross‐Validatory Choice and Assessment of Statistical Predictions , 1976 .

[18]  E. M. Wright,et al.  Adaptive Control Processes: A Guided Tour , 1961, The Mathematical Gazette.

[19]  M. Blumenstein,et al.  A modified direction feature for cursive character recognition , 2004, 2004 IEEE International Joint Conference on Neural Networks (IEEE Cat. No.04CH37541).

[20]  Francesco Camastra,et al.  Offline Cursive Character Challenge: a New Benchmark for Machine Learning and Pattern Recognition Algorithms. , 2006, 18th International Conference on Pattern Recognition (ICPR'06).

[21]  Hiroshi Yamada,et al.  Cursive handwritten word recognition using multiple segmentation determined by contour analysis , 1996 .

[22]  Gyeonghwan Kim,et al.  A Lexicon Driven Approach to Handwritten Word Recognition for Real-Time Applications , 1997, IEEE Trans. Pattern Anal. Mach. Intell..

[23]  Vladimir Vapnik,et al.  Statistical learning theory , 1998 .

[24]  Anthony J. Robinson,et al.  An Off-Line Cursive Handwriting Recognition System , 1998, IEEE Trans. Pattern Anal. Mach. Intell..

[26]  Sameer Singh,et al.  Cursive digit and character recognition in CEDAR database , 2000, Proceedings 15th International Conference on Pattern Recognition. ICPR-2000.

[27]  Nello Cristianini,et al.  Kernel Methods for Pattern Analysis , 2004 .

[28]  Teuvo Kohonen,et al.  Learning vector quantization , 1998 .

[29]  Ralf Herbrich,et al.  Learning Kernel Classifiers , 2001 .

[30]  Jung-Hsien Chiang,et al.  Handwritten word recognition with character and inter-character neural networks , 1997, IEEE Trans. Syst. Man Cybern. Part B.

[31]  Thorsten Joachims,et al.  Making large scale SVM learning practical , 1998 .

[32]  Teuvo Kohonen,et al.  Self-Organizing Maps , 2010 .

[33]  Brijesh Verma,et al.  An investigation of the modified direction feature for cursive character recognition , 2007, Pattern Recognit..

[34]  Stuart E. Dreyfus,et al.  Applied Dynamic Programming , 1965 .

[35]  J. Williamson.  Harmonic Analysis on Semigroups , 1967 .

[36]  Douglas A. Baxter,et al.  Small networks of empirically derived adaptive elements simulate some higher-order features of classical conditioning , 1990, Neural Networks.

[37]  Thomas Martinetz,et al.  'Neural-gas' network for vector quantization and its application to time-series prediction , 1993, IEEE Trans. Neural Networks.

[38]  Louis Vuurpijl,et al.  Finding structure in diversity: a hierarchical clustering method for the categorization of allographs in handwriting , 1997, Proceedings of the Fourth International Conference on Document Analysis and Recognition.

[39]  Moumita Ghosh,et al.  A novel approach for structural feature extraction: Contour vs. direction , 2004, Pattern Recognit. Lett..

[40]  Heekuck Oh,et al.  Neural Networks for Pattern Recognition , 1993, Adv. Comput..