Printed Thai Character Recognition Using Conditional Random Fields and Hierarchical Centroid Distance

Abstract:

This paper presents a Thai character recognition method based on topological properties. The method first extracts gradient features from a character image. A two-step classification are then applied to recognize the character. In the first step, a conditional random fields model is used to generate a set of possible characters. Then a nearest neighbor model based on hierarchical centroid distance is employed to finally recognize the character. The proposed method is trained by printed characters from documents and vehicle license plates. The technique is evaluated and found to have the recognition rate of 96.96%.

• Periodical: Applied Mechanics and Materials (Volumes 411-414)
• Pages: 1238-1246
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.411-414.1238
• Online since: September 2013
• Authors: Ousanee Sangkathum, Ohm Sornil
• Keywords: Conditional Random Fields, Gradient Feature Extraction, Hierarchical Centroid Distance, Thai Character Recognition

References

[1] B. B. Chaudhuri and U. Garain. Extraction of Type Style-based Meta-information from Imaged Documents, Int. J. on Document Analysis and Recognition, (2001).

DOI: 10.1007/pl00013557

[2] C. Sun, D. Si, Skew, and Slant. Correction for Document Images Using Gradient Direction. Proc. Int. Conf. On Document Analysis and Recognition (ICDAR), Vol. 1, 1997, pp.142-146.

DOI: 10.1109/icdar.1997.619830

[3] L. Zhang, Y. Lu and C. L. Tan. Italic Font Recognition Using Stroke Pattern Analysis on Wavelet Decomposed Word Images. Proceedings of the 17th International Conference on Pattern Recognition, Vol. 4, 2004, pp.835-838.

DOI: 10.1109/icpr.2004.1333902

[4] F. Kimura and M. Shridhar. Handwritten numerical recognition based on multiple algorithms, Pattern Recognition, Vol. 24, 1991, pp.969-983.

DOI: 10.1016/0031-3203(91)90094-l

[5] S. Srisuk. Thai printed character recognition using the Hausdorff distance. Proceedings of National Computer Science and Engineering Conference (NCSEC), (1999).

[6] A. Kawtrakul and P. Waewsawangwong. Multi-feature extraction for printed Thai character recognition. Proceeedings of the 4th Symposiumon Natural Language Processing (SNLP), (2000).

[7] R. Foopratheepsiri, W. Dejhan, and S. Rewthong. An improved accuracy method for Thai-OCR using fuzzy logic. Proceedings of the National Computer Science and Engineering Conference, (2004).

[8] P. Phokharakul and C. Kimpan. Recognition of handprinted Thai characters using the cavity features of character based on neural network. Proceedings of the 1998 IEEE Asia-Pacific Conference on Circuits and Systems, 1998, pp.149-152.

DOI: 10.1109/apccas.1998.743689

[9] B. Kijsirikul and S. Sinthupinyo. Approximate ILP rules by backpropagation neural network: a result on Thai character recognition. Springer-Verlag Berin Heidelberg, 1999, pp.162-173.

DOI: 10.1007/3-540-48751-4_16

[10] A. Thammano and P. Duangphasuk. Printed Thai character recognition using hierarchical cross-correlation ARTMAP. Proceedings of the 17th IEEE International Conference on Tools with Artificial Intelligence, (2005).

DOI: 10.1109/ictai.2005.100

[11] S. Mitatha, K. Dejharn, F. Chevasuvit, B. Chankuang and W. Kasemsiri. Experimental results of using rough sets of printed Thai characters recognition. Proceedings of IEEE Region 10th International Conference on Electrical and Electronic Technology, 2001, pp.331-334.

DOI: 10.1109/tencon.2001.949608

[12] S. Mitatha, K. Dejharn, F. Chevasuvit and W. Kasemsiri. Some experimental results of using rough sets for printed Thai characters recognition. International Journal of Computational Cognition, Vol. 1, 2003, pp.109-121.

DOI: 10.1109/tencon.2001.949608

[13] C. Pornpanomchai and M. Daveloh. Printed Thai character recognition by genetic algorithm. Proceedings of the 6th Conference on Machine Learning and Cybernetics, 2007, pp.3354-3359.

DOI: 10.1109/icmlc.2007.4370727

[14] S. Tangwongsan and O. Jungthanawong. A refinement of stroke structure for printed Thai character recognition. Proceeedings of the 9th International Conference on Signal Processing, 2008, pp.1504-1507.

DOI: 10.1109/icosp.2008.4697418

[15] C. H. Bhabhor and A. B. Upadhyay. Performance Analysis for Image Segmentation of Various EDGE detection techniques. International Conference on Research & Development in Engineering, Technology & Sciences (ICRD), 2013, p.401 – 402.

[16] J. Canny. A computational approach to edge detection. Pattern Analysis and Machine Intelligence, IEEE Transactions on PAMI, 1986, p.679–698.

DOI: 10.1109/tpami.1986.4767851

[17] M. Chen, Y. Chen , M. R. Brent, and A. E. Tenny. Gradient-Based Feature Selection for Conditional Random Fields and Its Applications in Computational Genetics. IEEE 21st International Conference on Tools with Artificial Intelligence (ICTAI), 2009, pp.750-757.

DOI: 10.1109/ictai.2009.82

[18] S. Armon. Handwriting Recognition and Fast Retrieval for Hebrew Historical Manuscripts. Hewbrew University, 2011, pp.17-18.

[19] A. Sexton, A. Todman, and K. Woodward. Font Recognition using shape-based quad-tree and kd-tree decomposition, JCIS, 2000, pp.212-21.