Advertisement

Adaptive Histogram of Oriented Gradient for Printed Thai Character Recognition

  • Kuntpong Woraratpanya
  • Taravichet Titijaroonroj
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 265)

Abstract

A similarity of printed Thai characters is a grand challenge of optical character recognition (OCR), especially in case of a variety of font types, sizes, and styles. This paper proposes an effective feature extraction, adaptive histogram of oriented gradient (AHOG), for overcoming the character similarity. The proposed method improves the conventional histogram of oriented gradient (HOG) in two principal phases, which are (i) adaptive partition for gradient images and (ii) adaptive binning for oriented histograms. The former is implemented with quadtree partition based on gradient image variance so as to provide for an effective local feature extraction. The later is implemented with non-uniform mapping technique, so that the AHOG descriptor can be constructed with minimal errors. Based on 59,408 single character images equally divided into training and testing samples, the experimental results show that the AHOG method outperforms the conventional HOG and state-of-the-art methods, including scale space histogram of oriented gradient (SSHOG), pyramid histogram of oriented gradient (PHOG), multilevel histogram of oriented gradient (MHOG), and HOG column encoding algorithm (HOG-Column).

Keywords

Printed Thai Character Recognition Pattern Recognition Histogram of Oriented Gradient (HOG) Adaptive Histogram of Oriented Gradient (AHOG) Feature Extraction 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Kimpan, C., Itoh, A., Kawanishi, K.: Fine Classification of Printed Thai Character Recognition Using the Karhunen-Loeve Expansion. IEE Proceedings 134, 257–264 (1987)Google Scholar
  2. 2.
    Tanprasert, C., Sae-Tang, S.: Thai type style recognition. In: 1999 IEEE International Conference on Circuits and Systems, pp. 336–339. IEEE Press, New York (1999)Google Scholar
  3. 3.
    Thammano, A., Duangphasuk, P.: Printed Thai Character Recognition Using the Hierarchical Cross-correlation ARTMAP. In: The 17th IEEE International Conference on Tools with Artificial Intelligence, pp. 695–698. IEEE Press, New York (2005)Google Scholar
  4. 4.
    Woraratpanya, K., Titijaroonrog, T.: Printed Thai Character Recognition Using Standard Descriptor. In: Meesad, P., Unger, H., Boonkrong, S. (eds.) IC2IT2013. AISC, vol. 209, pp. 165–173. Springer, Heidelberg (2013)CrossRefGoogle Scholar
  5. 5.
    Navneet, D., Bill, T.: Histograms of Oriented Gradients for Human Detection. In: IEEE International Conference on Computer Vision and Pattern Recognition, pp. 1–8. IEEE Press, New York (2005)Google Scholar
  6. 6.
    Ning, H., Jiaheng, C., Lin, S.: Scale Space Histogram of Oriented Gradients for Human Detection. In: 2008 International Symposium on Information Science and Engineering, pp. 167–170. IEEE Press, New York (2008)CrossRefGoogle Scholar
  7. 7.
    Anna, B., Andrew, Z., Xavier, M.: Representing Shape with a Spatial Pyramid Kernel. In: The 6th ACM International Conference on Image and Video Retrieval (2007)Google Scholar
  8. 8.
    Subhransu, M., Alexander, C.B., Jitendra, M.: Classification Using Intersection Kernel Support Vector Machines is Efficient. In: IEEE International Conference on Computer Vision and Pattern Recognition, IEEE Press, New York (2008)Google Scholar
  9. 9.
    Andrew, J.N., Lewis, D.G.: Multiscale Histogram of Oriented Gradient Descriptors for Robust Character Recognition. In: 2011 International Conference on Document Analysis and Recognition, pp. 1085–1089. IEEE Press, New York (2011)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  • Kuntpong Woraratpanya
    • 1
  • Taravichet Titijaroonroj
    • 1
  1. 1.Faculty of Information TechnologyKing Mongkut’s Institute of Technology LadkrabangBangkokThailand

Personalised recommendations