This Article 
 Bibliographic References 
 Add to: 
Character Recognition Without Segmentation
September 1995 (vol. 17 no. 9)
pp. 903-909

Abstract—A segmentation-free approach to OCR is presented as part of a knowledge-based word interpretation model. This new method is based on the recognition of subgraphs homeomorphic to previously defined prototypes of characters [16]. Gaps are identified as potential parts of characters by implementing a variant of the notion of relative neighborhood used in computational perception. In the system, each subgraph of strokes that matches a previously defined character prototype is recognized anywhere in the word even if it corresponds to a broken character or to a character touching another one. The characters are detected in the order defined by the matching quality. Each subgraph that is recognized is introduced as a node in a directed net that compiles different alternatives of interpretation of the features in the feature graph. A path in the net represents a consistent succession of characters in the word. The method allows the recognition of characters that overlap or that are underlined. A final search for the optimal path under certain criteria gives the best interpretation of the word features. The character recognizer uses a flexible matching between the features and a flexible grouping of the individual features to be matched. Broken characters are recognized be looking for gaps between features that may be interpreted as part of a character. Touching characters are recognized because the matching allows nonmatched adjacent strokes. The recognition results of this system for over 24,000 printed numeral characters belonging to a USPS database and on some hand-printed words confirmed the method’s high robustness level.

[1] M. Berthod and S. Ahyan,“On-line cursive script recognition: A structural approach,” Proc. Fifth Int’l Conf. Pattern Recognition, pp. 723-725, 1980.
[2] R.G. Casey and G. Nagy,“Recursive segmentation and classification of composite character patterns,” Proc. Sixth Int’l Conf. Pattern Recognition, pp. 1,023-1,026,Munich, 1982.
[3] S. Edelman, T. Flash, and S. Ullman, “Reading Cursive Handwriting by Alignment of Letter Prototypes,” Int'l J. Computer Vision, vol. 5, no. 3, pp. 303-331, 1990.
[4] E. Edelsbruner,D.G. Kirkpatrick,, and R. Siegel,“On the shape of a set of points in the plane,” IEEE Trans. Information Theory, vol. 29, no. 4, pp. 551-559, 1983.
[5] H. Fujisawa,Y. Nakano,, and K. Kurino,“Segmentation methods for character recognition: From segmentation to document structure analysis,” IEEE Proc., pp. 1,079-1,092, July 1992.
[6] M. Gilloux,“Hidden Markov models in handwriting recognition,” S. Impedovo, ed., Fundamentals of Handwriting Recognition, vol. 124of NATOASI Series F, pp. 264-288. Springer-Verlag, 1994.
[7] V. Govindaraju and S.N. Srihari,“Separating handwritten text from interfering strokes,” personal communication, 1992.
[8] J. Hull,S. Srihari,, and R. Choudhari,“An integrated algorithm for text recognition: Comparison with a cascaded algorithm,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 5, no. 4, pp. 384-395, 1983.
[9] J. Hu, B. Sakoda, and T. Pavlidis, "Interactive road finding for aerial images," Proc. IEEE Workshop Applications of Computer Vision, pp. 56-63, 1992.
[10] D.B. Kirkpatrick and J.D. Radke,“A framework for computational geometry,” G.T. Toussaint, ed., Computational Geometry, pp. 217-248.North Holland, 1985.
[11] E. Lecolinet and J. Crettez,“A grapheme-based segmentation technique for cursive script recognition,” First Int’l Conf. Document Analysis and Recognition, pp. 740-748,Saint-Marlo, France, Sept. 1991.
[12] D.G. Lowe, Perceptual Organization and Visual Recognition. Boston: Kluwer Academic, 1985.
[13] K. Ohmori,“On-line handwritten kanji character recognition using hypothesis generation in the space of hierarchical knowledge,” Proc. Third Int’l Workshop Frontiers in Handwriting Recognition, pp. 242-251,Buffalo, N.Y., May 1993.
[14] M. Okamoto and B. Miao,“Recognition of mathematical expressions by using the layout structures of symbols,” First Int’l Conf. Document Analysis and Recognition, pp. 242-250,Saint-Marlo, France, Sept. 1991.
[15] J.D. Radke,“On the shape of a set of points,” G.T. Toussaint, ed., Computational Morphology, pp. 105-136.North-Holland, 1988.
[16] J. Rocha and T. Pavlidis, "A Shape Analysis Model With Applications to a Character Recognition System," IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 16, pp. 393-404, 1994.
[17] J. Schümann,N. Bartneck,T. Bayer,J. Franke,E. Mandler,, and M. Oberländer,“Document analysis—from pixels to contents,” IEEE Proc., vol. 80, pp. 1,101-1,119, July 1992.
[18] K.L. Supowit,“The relative neighborhood graph with an application to minimum spanning trees,” J. ACM, vol. 30, pp. 428-448, 1983.
[19] W.J. Sakoda,J. Zhou,, and T. Pavlidis,“Feature extraction directly from gray scale with applications to an address recognition system,” SPIE Symp. Electronic Imaging and Technology, pp. 21-29,San Jose, Calif., Feb. 1993.
[20] S. Tsujimoto and H. Asada,“Resolving ambiguity in segmenting touching characters,” First Int’l Conf. Document Analysis and Recognition, pp. 701-709,Saint-Marlo, France, Sept. 1991.
[21] G.T. Toussaint,“The relative neighborhood graph of a finite planar set,” Pattern Recognition, vol. 12, pp. 261-268, 1980.
[22] G.T. Toussaint,“A graph-theoretical primal sketch,” G.T. Toussaint, ed., Computational Morphology, pp. 229-260.North-Holland, 1988.
[23] J.I. Toriwaki and S. Yokoi,“Voronoi and related neighbors on digitized two dimensional space with application to texture analysis,” G.T. Toussaint, ed., Computational Morphology, pp. 207-228,North-Holland, 1988.

Index Terms:
Character recognition without segmentation, broken character recognition, touching character recognition, homeomorphic subgraph matching, relative neighborhood graph.
Jairo Rocha, Theo Pavlidis, "Character Recognition Without Segmentation," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 17, no. 9, pp. 903-909, Sept. 1995, doi:10.1109/34.406657
Usage of this product signifies your acceptance of the Terms of Use.