Convolutional Neural Networks for Finance Image Classification

  • Xingjie Zhu
  • Yan LiuEmail author
  • Xingwang Liu
  • Chi Li
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 849)


In recent years, deep convolutional neural networks have demonstrated excellent performance on visual tasks, such as image classification. Based on the deep architecture, this paper designs a new method to handle an automatic financial recognition problem, namely to recognize the value-added tax (VAT) invoices, finance-related documents, and other finance-related images. The proposed method consists of three steps: first, image preprocessing will be performed on the original image and the augmented image will be separated into four patches for further processing; thus the obtained image patches will be the input of a deep convolutional neural model for the training purpose; at the final step, we use the four predications which obtained from the previous step to determine the final categorizes. Our experimental result shows that this method can conduct finance image classification with high performance.


Image classification Segmentation Neural networks 


  1. 1.
    Milgram, J., Sabourin, R., Cheriet, M.: Two-stage classification system combining model-based and discriminative approaches. In: The 17th International Conference on Pattern Recognition, pp. 152–155 (2004)Google Scholar
  2. 2.
    Farid, M., Lorenzo, B.: Classification of hyperspectral remote sensing images with support vector machines. IEEE Trans. Geosci. Remote Sens. 42(8), 1778–1790 (2004)CrossRefGoogle Scholar
  3. 3.
    Xu, Y., Zhang, B., Zhong, Z.: Multiple representations and sparse representation for image classification. Pattern Recogn. Lett. 68, 9–14 (2015)CrossRefGoogle Scholar
  4. 4.
    Guo, S., Huang, W., Qiao, Y.: Improving scale invariant feature transform with local color contrastive descriptor for image classification. J. Electron. Imaging 26(1), 013015 (2017)CrossRefGoogle Scholar
  5. 5.
    Guo, Z.H., Zhang, L., Zhang, D.: A completed modeling of local binary pattern operator for texture classification. IEEE Trans. Image Process. 19(6), 1657–1663 (2010)MathSciNetCrossRefGoogle Scholar
  6. 6.
    Zhou, L., Wang, L., Liu, L., Ogunbona, P., Shen, D.: Support vector machines for neuroimage analysis: interpretation from discrimination. In: Support Vector Machines Applications, pp. 191–220. Springer, Cham (2014)Google Scholar
  7. 7.
    Sun, G., Chen, T., Su, Y., et al.: Internet traffic classification based on incremental support vector machines. Mobile Netw. Appl. 23(120), 1–8 (2018)Google Scholar
  8. 8.
    Xu, Y., Zhu, Q., Fan, Z., Zhang, D., Mi, J., Lai, Z.: Using the idea of the sparse representation to perform coarse-to-fine face recognition. Inf. Sci. 238(20), 138–148 (2013)MathSciNetCrossRefGoogle Scholar
  9. 9.
    Rawat, W., Wang, Z.: Deep convolutional neural networks for image classification: a comprehensive review. Neural Comput. 29(9), 1 (2017)CrossRefGoogle Scholar
  10. 10.
    Williams, T., Li, R.: An ensemble of convolutional neural networks using wavelets for image classification. J. Softw. Eng. Appl. 11(2), 69–88 (2018)CrossRefGoogle Scholar
  11. 11.
    Sharma, A., Liu, X., Yang, X., et al.: A patch-based convolutional neural network for remote sensing image classification. Neural Netw. 95, 19–28 (2017)CrossRefGoogle Scholar
  12. 12.
    Steinkrau, D., Simard, P.Y., Buck, I.: Using GPUs for machine learning algorithms. In: Proceedings of the 8th International Conference on Document Analysis and Recognition, Washington, pp. 1115–1119 (2005)Google Scholar
  13. 13.
    Hinton, G.E., Osindero, S., Teh, Y.: A fast learning algorithm for deep belief nets. Neural Comput. 18(7), 1527–1554 (2006)MathSciNetCrossRefGoogle Scholar
  14. 14.
    Bhargava, D.N., Kumawat, A., Bhargava, D.R.: Threshold and binarization for document image analysis using otsu’s Algorithm. Int. J. Comput. Trends Technol. 17(5), 272–275 (2014)CrossRefGoogle Scholar
  15. 15.
    Xia, X., Xu, C., Nan, B.: Inception-v3 for flower classification. In: 2017 2nd International Conference on Image, Vision and Computing (ICIVC), Chengdu, pp. 783–787 (2017)Google Scholar
  16. 16.
    Simon, M., Rodner, E., Denzler, J.: ImageNet pre-trained models with batch normalization, pp. 1–4 (2016)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.Taikang Insurance GroupBeijingChina

Personalised recommendations