Segmentation-based recognition of handwritten touching pairs of digits using structural features

In this paper, we propose a segmentation-based recognition method of handwritten touching pairs of digits using structural features of contour. Four kinds of candidate break points are obtained from contour and six touching types are defined based on an analysis of the ligature and the characteristics of candidate break points. The final break points of touching pairs of digits are deduced by verifying candidate segment combinations. The main advantages of this method are that reliable segment combinations are used in the multiple hypothesis recognition, and segmentation error of traditional segmentation-based recognition method are reduced by verifying segment combinations. To evaluate the proposed method, we have experimented with 3500 touching pairs of digits of the NIST SD19 database. An encouraging recognition rate of 92.5% has been obtained.

[1]  Seong-Whan Lee,et al.  Integrated segmentation and recognition of connected handwritten characters with recurrent neural network , 1995, Proceedings of 3rd International Conference on Document Analysis and Recognition.

[2]  Pengfei Shi,et al.  A background-thinning-based approach for separating and recognizing connected handwritten digit strings , 1999, Pattern Recognit..

[3]  Shaowei Xia,et al.  A new method for segmenting unconstrained handwritten numeral string , 1997, Proceedings of the Fourth International Conference on Document Analysis and Recognition.

[4]  Hirobumi Nishida,et al.  A Model-Based Split-And-Merge Method for Character String Recognition , 1994, Int. J. Pattern Recognit. Artif. Intell..

[5]  Yi Lu,et al.  Character segmentation in handwritten words - An overview , 1996, Pattern Recognit..

[6]  E. Lecolinet,et al.  Strategies in character segmentation: a survey , 1995, Proceedings of 3rd International Conference on Document Analysis and Recognition.

[7]  Gyeonghwan Kim,et al.  Chaincode Contour Processing for Handwritten Word Recognition , 1999, IEEE Trans. Pattern Anal. Mach. Intell..

[8]  Azriel Rosenfeld,et al.  From pixels to features: J C Simon (ed). Published by Nort-Holland, Netherlands. 1989. 416 pp. $94.75 , 1990 .

[9]  Keiji Yamada,et al.  A New Numeral String Recognition Method Using Character Touching Type Verification , 1999 .

[10]  Hong Yan,et al.  Construction of partitioning paths for touching handwritten characters , 1999, Pattern Recognit. Lett..

[11]  Bart Kosko,et al.  Neural networks and fuzzy systems , 1998 .

[12]  Venu Govindaraju,et al.  Holistic recognition of touching digits , 1998 .

[13]  Yasuaki Nakano,et al.  Segmentation methods for character recognition: from segmentation to document structure analysis , 1992, Proc. IEEE.

[14]  Venu Govindaraju,et al.  Segmentation and recognition of connected handwritten numeral strings , 1997, Pattern Recognit..

[15]  James D. Keeler,et al.  A Self-Organizing Integrated Segmentation and Recognition Neural Net , 1991, NIPS.

[16]  Alexander Filatov,et al.  Graph-based handwritten digit string recognition , 1995, Proceedings of 3rd International Conference on Document Analysis and Recognition.

[17]  Horst Bunke,et al.  Off-line handwritten numeral string recognition by combining segmentation-based and segmentation-free methods , 1998, Pattern Recognit..

[18]  Adam Krzyzak,et al.  Segmentation of handwritten digits using contour features , 1993, Proceedings of 2nd International Conference on Document Analysis and Recognition (ICDAR '93).