Graph-matching-based character recognition for Chinese seal images

Abstract

Recognizing characters in Chinese seal images is important when researching ancient cultural artworks because the seals may contain critical historical information. However, owing to large intraclass variance and a limited number of training samples, recognizing such characters in Chinese seals is challenging. Thus, this study proposes a graph-matching-based method to recognize characters in historical Chinese seal images. In the proposed method, a Chinese seal character is first modeled as a graph representing its underlying geometric structure. Then, two affinity matrices that measure the similarity of nodes and edge pairs are calculated with their local features. Finally, a correspondence matrix is calculated using a graph matching algorithm and the most similar reference is selected as the recognition result. Compared with several existing classification methods for seal image recognition, the proposed graph-matching-based method achieves better results, particularly in the case of limited samples.

This is a preview of subscription content, access via your institution.

References

  1. 1

    Roy P P, Pal U, Lladós J. Document seal detection using GHT and character proximity graphs. Pattern Recogn, 2011, 44: 1282–1295

    Article  Google Scholar 

  2. 2

    Ren C, Liu D, Chen Y B. A new method on the segmentation and recognition of Chinese characters for automatic Chinese seal imprint retrieval. In: Proceedings of International Conference on Document Analysis and Recognition, 2011. 972–976

  3. 3

    Roy P P, Pal U, Lladós J. Seal detection and recognition: an approach for document indexing. In: Proceedings of International Conference on Document Analysis and Recognition, 2009. 101–105

  4. 4

    Yin F, Wang Q F, Zhang X Y, et al. Chinese handwriting recognition competition. In: Proceedings of International Conference on Document Analysis and Recognition, 2013. 1464–1469

  5. 5

    Wang C H, Xiao B H, Dai R W. Parallel compact integration in handwritten Chinese character recognition. Sci China Ser F-Inf Sci, 2004, 47: 89–96

    Article  Google Scholar 

  6. 6

    Guo J, Wang C H, Roman-Rangel E, et al. Building hierarchical representations for Oracle character and sketch recognition. IEEE Trans Image Process, 2016, 25: 104–118

    MathSciNet  Article  MATH  Google Scholar 

  7. 7

    Sebastian T B, Klein P N, Kimia B B. Recognition of shapes by editing their shock graphs. In: Proceedings of IEEE International Conference on Computer Vision, 2011. 755–762

  8. 8

    Bai X, Latecki L J. Path similarity skeleton graph matching. IEEE Trans Pattern Anal Mach Intell, 2008, 30: 1282–1292

    Article  Google Scholar 

  9. 9

    Zhang H, Mu Y, You Y H, et al. Multi-scale sparse feature point correspondence by graph cuts. Sci China Inf Sci, 2010, 53: 1224–1232

    MathSciNet  Article  Google Scholar 

  10. 10

    Zhang L M, Yang Y, Wang M, et al. Detecting densely distributed graph patterns for fine-grained image categorization. IEEE Trans Image Process, 2016, 25: 553–565

    MathSciNet  Article  MATH  Google Scholar 

  11. 11

    Mian A S, Bennamoun M, Owens R. Three-dimensional model-based object recognition and segmentation in cluttered scenes. IEEE Trans Pattern Anal Mach Intell, 2006, 28: 1584–1601

    Article  Google Scholar 

  12. 12

    Jiang H, Yu S X, Martin D R. Linear scale and rotation invariant matching. IEEE Trans Pattern Anal Mach Intell, 2011, 33: 1339–1355

    Article  Google Scholar 

  13. 13

    Zhao R, Martinez A M. Labeled graph kernel for behavior analysis. IEEE Trans Pattern Anal Mach Intell, 2016, 38: 1640–1650

    Article  Google Scholar 

  14. 14

    Aksoy E E, Abramov A, Worgotter F, et al. Categorizing object-action relations from semantic scene graphs. In: Proceedings of IEEE International Conference on Robotics and Automation, 2010. 398–405

  15. 15

    Belongie S, Malik J. Matching with shape contexts. In: Proceedings Workshop on Content-based Access of Image and Video Libraries, 2000

  16. 16

    Liu C L, Kim I J, Kim J H. Model-based stroke extraction and matching for handwritten Chinese character recognition. Pattern Recogn, 2001, 34: 2339–2352

    Article  MATH  Google Scholar 

  17. 17

    Fischler M A, Bolles R C. Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography. Read Comput Vision, 1987, 24: 726–740

    Google Scholar 

  18. 18

    Schenker P S. Method for registration of 3-D shapes. IEEE Trans Pattern Anal Mach Intell, 2002, 14: 239–256

    Google Scholar 

  19. 19

    Cho M, Lee J, Lee K M. Reweighted random walks for graph matching. In: Proceedings European Conference on Computer Vision, 2010. 492–505

  20. 20

    Mateus D, Horaud R, Knossow D, et al. Articulated shape matching using Laplacian eigenfunctions and unsupervised point registration. In: Proceedings IEEE Conference on Computer Vision and Pattern Recognition, 2008

  21. 21

    Zhou F, De la Torre F. Factorized graph matching. IEEE Trans Pattern Anal Mach Intell, 2016, 38: 1774–1789

    Article  Google Scholar 

  22. 22

    Otsu N. A threshold selection method from gray-level histograms. IEEE Trans Syst Man Cybern, 1979, 9: 62–66

    Article  Google Scholar 

  23. 23

    Wang Y, Zhong B J. A scale-space technique for polygonal approximation of planar curves. In: Proceedings of IEEE International Conference on Image Processing, 2013. 517–520

  24. 24

    Fukushima M. A modified Frank-Wolfe algorithm for solving the traffic assignment problem. Transpation Res Part B-Meth, 1984, 18: 169–177

    MathSciNet  Article  Google Scholar 

  25. 25

    Chang C C, Lin C J. LIBSVM: a library for support vector machines. ACM Trans Intell Syst Technol, 2011, 2: 1–27

    Article  Google Scholar 

  26. 26

    Lecun Y, Bottou L, Bengio Y, et al. Gradient-based learning applied to document recognition. Proc IEEE, 1998, 86: 2278–2324

    Article  Google Scholar 

  27. 27

    You C, Robinson D P, Vidal R. Scalable sparse subspace clustering by orthogonal matching pursuit. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 2016. 3918–3927

  28. 28

    Qu X W, Wang W Q, Lu K. In-air handwritten Chinese character recognition using discriminative projection based on locality-sensitive sparse representation. In: Proceedings of International Conference on Pattern Recognition, 2017. 1137–1140

  29. 29

    Cao J L, Pang Y W, Li X L, et al. Randomly translational activation inspired by the input distributions of ReLU. Neurocomputing, 2018, 275: 859–868

    Article  Google Scholar 

Download references

Acknowledgements

This work was supported by National Natural Science Foundation of China (Grant Nos. 6152010-6001, 61801178), Natural Science Foundation of Hunan Province (Grant No. 2018JJ3071), and by Hunan Key Laboratory of Visual Perception and Artificial Intelligence.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Shutao Li.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Sun, B., Hua, S., Li, S. et al. Graph-matching-based character recognition for Chinese seal images. Sci. China Inf. Sci. 62, 192102 (2019). https://doi.org/10.1007/s11432-018-9724-7

Download citation

Keywords

  • graph model
  • Chinese seal
  • character recognition
  • graph matching