Abstract
In our previous work, without considering the stroke information, a method based on spectral clustering (SC) for solving handwritten touching numerals segmentation was proposed and obtained very good performance. In this paper, we extend the algorithm to an online system, and propose an improved method where the stroke information is involved. First, the features of the numerals image are extracted by a sliding window. Second, the obtained feature vectors are trained by support vector machine to generate an affinity matrix. Thereafter, the stroke information of original images is used to generate another affinity matrix. Finally, these two affinity matrices are added and trained by SC. Experimental results show that the proposed method can further improve the accuracy of segmentation.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Chen, C., Guo, J.: A general approach for handwritten digits segmentation using spectral clustering. In: International Conference on Document Analysis and Recognition (IAPR), pp. 547–552 (2018)
Chen, Y.K., Wang, J.F.: Segmentation of single or multiple-touching handwritten numeral string using background and foreground analysis. IEEE Pattern Anal. 1, 1304–1317 (2000)
Congedo, G., Dimauro, G., Impedovo, S., Pirlo, G.: Segmentation of numeric strings. In: International Conference on Document Analysis and Recognition, vol. 2, pp. 1038–1041 (1995)
Corr, P.J., Silvestre, G.C., Bleakley, C.J.: Open source dataset and deep learning models for online digit gesture recognition on touchscreens. In: Irish Machine Vision and Image Processing Conference (IMVIP) (2017)
Diem, M., Fiel, S., Garz, A., Keglevic, M., Kleber, F., Sablatnig, R.: ICDAR 2013 competition on handwritten digit recognition (HDRC 2013). In: International Conference on Document Analysis and Recognition, pp. 1422–1427 (2013)
Diem, M., et al.: ICFHR 2014 competition on handwritten digit string recognition in challenging datasets (HDSRC 2014). In: International Conference on Frontiers in Handwriting Recognition, pp. 779–784 (2014)
Elnagar, A., Alhajj, R.: Segmentation of connected handwritten numeral strings. Pattern Recogn. 36(3), 625–634 (2003)
Fujisawa, H., Nakano, Y., Kurino, K.: Segmentation methods for character recognition: from segmentation to document structure analysis. Proc. IEEE 80(7), 1079–1092 (1992)
Graves, A.: Offline arabic handwriting recognition with multidimensional recurrent neural networks. In: Advances in Neural Information Processing Systems, pp. 549–558 (2008)
Grother, P.J.: Nist special database 19, Handprinted forms and characters database. National Institute of Standards and Technology (NIST) (1995)
Kim, K.K., Jin, H.K., Suen, C.Y.: Segmentation-based recognition of handwritten touching pairs of digits using structural features. Pattern Recogn. Lett. 23, 13–24 (2002)
Lei, Y., Liu, C.S., Ding, X.Q., Fu, Q.: A recognition based system for segmentation of touching handwritten numeral strings. In: International Workshop on Frontiers in Handwriting Recognition, pp. 294–299. IEEE (2004)
Lu, Z., Chi, Z., Siu, W.C., Shi, P.: A background-thinning-based approach for separating and recognizing connected handwritten digit strings. Pattern Recogn. 32(6), 921–933 (1999)
Oliveira, L.S., Lethelier, E., Bortolozzi, F.: A new approach to segment handwritten digits. In: Proceedings of 7th International Workshop on Frontiers in Handwriting Recognition, pp. 577–582 (2000)
Pal, U., Choisy, C.: Touching numeral segmentation using water reservoir concept. Pattern Recogn. Lett. 24(1), 261–272 (2003)
Ribas, F.C., Oliveira, L.S., Britto, A.S., Sabourin, R.: Handwritten digit segmentation: a comparative study. Int. J. Doc. Anal. Recogn. 16(2), 127–137 (2013)
Sadri, J., Suen, C.Y., Bui, T.D.: A genetic framework using contextual knowledge for segmentation and recognition of handwritten numeral strings. Pattern Recogn. 40(3), 898–919 (2007)
Shi, Z., Govindaraju, V.: Segmentation and recognition of connected handwritten numeral strings. Pattern Recogn. 30(9), 1501–1504 (1997)
Suwa, M., Naoi, S.: Segmentation of handwritten numerals by graph representation. In: International Workshop on Frontiers in Handwriting Recognition, vol. 2, pp. 334–339 (2004)
Vellasques, E., Oliveira, L.S., Koerich, A.L., Sabourin, R.: Filtering segmentation cuts for digit string recognition. Pattern Recogn. 41(10), 3044–3053 (2008)
Yu, D., Yan, H.: Separation of touching handwritten multi-numeral strings based on morphological structural features. Pattern Recogn. 34(3), 587–599 (2001)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer Nature Switzerland AG
About this paper
Cite this paper
Shao, R., Chen, C., Guo, J. (2018). An Online Handwritten Numerals Segmentation Algorithm Based on Spectral Clustering. In: Cheng, L., Leung, A., Ozawa, S. (eds) Neural Information Processing. ICONIP 2018. Lecture Notes in Computer Science(), vol 11305. Springer, Cham. https://doi.org/10.1007/978-3-030-04221-9_45
Download citation
DOI: https://doi.org/10.1007/978-3-030-04221-9_45
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-04220-2
Online ISBN: 978-3-030-04221-9
eBook Packages: Computer ScienceComputer Science (R0)