Ink Analysis Using CNN-Based Transfer Learning to Detect Alteration in Handwritten Words

Conference paper
Part of the Communications in Computer and Information Science book series (CCIS, volume 1148)


Alteration of words in handwritten financial documents such as cheques, medical claims, and insurance claims may lead to monetary loss to the customers and financial institutions. Hence, automatic identification of such alteration in documents is a crucial task. Therefore, an ink color based analysis using Convolutional Neural Network (CNN) automation method has been introduced for alteration detection. Pre-trained AlexNet and VGG-16 architectures have been used to study the effect of transfer learning on the problem at hand. Further, two different shallow CNNs have been employed for recognition. A data set has been created using ten blue and ten black pens to simulate the word alteration problem. The dataset captures the word alteration by addition of the characters (or even pen strokes) in the existing word. Experiments have revealed that the transfer learning based deep CNN architectures have outperformed the shallow CNN architectures on both blue and black pens.


Convolutional Neural Network Document alteration Handwritten forensics Transfer learning 


  1. 1.
  2. 2.
    Boufenar, C., Kerboua, A., Batouche, M.: Investigation on deep learning for off-line handwritten Arabic character recognition. Cognit. Syst. Res. 50, 180–195 (2018)CrossRefGoogle Scholar
  3. 3.
    Chhabra, S., Gupta, G., Gupta, M., Gupta, G.: Detecting fraudulent bank checks. In: Advances in Digital Forensics XIII, pp. 245–266 (2017)Google Scholar
  4. 4.
    Dansena, P., Bag, S., Pal, R.: Differentiating pen inks in handwritten bank cheques using multi-layer perceptron. In: International Conference on Pattern Recognition and Machine Intelligence, pp. 655–663 (2017)Google Scholar
  5. 5.
    Dasari, H., Bhagvati, C.: Identification of non-black inks using HSV colour space. In: International Conference on Document Analysis and Recognition, pp. 486–490 (2007)Google Scholar
  6. 6.
    Deng, J., Dong, W., Socher, R., Li, L.J., Li, K., Fei-Fei, L.: ImageNet: a large-scale hierarchical image database. In: International Conference on Computer Vision and Pattern Recognition, pp. 248–255 (2009)Google Scholar
  7. 7.
    Gorai, A., Pal, R., Gupta, P.: Document fraud detection by ink analysis using texture features and histogram matching. In: International Joint Conference on Neural Networks, pp. 4512–4517 (2016)Google Scholar
  8. 8.
    Harris, J.: Developments in the analysis of writing inks on questioned documents. J. Forensic Sci. 37(2), 612–619 (1992)CrossRefGoogle Scholar
  9. 9.
    Hollander, M., Wolfe, D.A., Chicken, E.: Nonparametric statistical methods, vol. 751. Wiley, Hoboken (2013)zbMATHGoogle Scholar
  10. 10.
    Krizhevsky, A., Sutskever, I., Hinton, G.E.: Imagenet classification with deep convolutional neural networks. In: Advances in Neural Information Processing Systems, pp. 1097–1105 (2012)Google Scholar
  11. 11.
    Kumar, R., Pal, N.R., Chanda, B., Sharma, J.: Forensic detection of fraudulent alteration in ball-point pen strokes. IEEE Trans. Inf. Forensics Secur. 7(2), 809–820 (2012)CrossRefGoogle Scholar
  12. 12.
    Kumar, R., Pal, N.R., Sharma, J.D., Chanda, B.: A novel approach for detection of alteration in ball pen writings. In: International Conference on Pattern Recognition and Machine Intelligence, pp. 400–405 (2009)Google Scholar
  13. 13.
    LeCun, Y.A., Bottou, L., Orr, G.B., Müller, K.R.: Efficient backprop. In: Neural Networks: Tricks of the Trade, pp. 9–48 (2012)Google Scholar
  14. 14.
    Lei, L., Zhu, H., Gong, Y., Cheng, Q.: A deep residual networks classification algorithm of fetal heart CT images. In: International Conference on Imaging Systems and Techniques, pp. 1–4 (2018)Google Scholar
  15. 15.
    Liu, X., Wang, C., Hu, Y., Zeng, Z., Bai, J., Liao, G.: Transfer learning with convolutional neural network for early gastric cancer classification on magnifiying narrow-band imaging images. In: International Conference on Image Processing, pp. 1388–1392 (2018)Google Scholar
  16. 16.
    Meyer, B.J., Harwood, B., Drummond, T.: Deep metric learning and image classification with nearest neighbour gaussian kernels. In: International Conference on Image Processing, pp. 151–155 (2018)Google Scholar
  17. 17.
    Pereira, J.F.Q., et al.: Projection pursuit and PCA associated with near and middle infrared hyperspectral images to investigate forensic cases of fraudulent documents. Microchem. J. 130, 412–419 (2017)CrossRefGoogle Scholar
  18. 18.
    Pramanik, R., Dansena, P., Bag, S.: A study on the effect of CNN-based transfer learning on handwritten Indic and mixed numeral recognition. In: Workshop on Document Analysis and Recognition, pp. 41–51 (2018)Google Scholar
  19. 19.
    Roy, S., Sangineto, E., Sebe, N., Demir, B.: Semantic-fusion gans for semi-supervised satellite image classification. In: International Conference on Image Processing, pp. 684–688 (2018)Google Scholar
  20. 20.
    Simonyan, K., Zisserman, A.: Very deep convolutional networks for large-scale image recognition (2014). arXiv preprint arXiv:1409.1556
  21. 21.
    Suzuki, M., Akiba, N., Kurosawa, K., Akao, Y., Higashikawa, Y.: Differentiation of black writing ink on paper using luminescence lifetime by time-resolved luminescence spectroscopy. Forensic Sci. Int. 279, 281–287 (2017)CrossRefGoogle Scholar
  22. 22.
    Too, E.C., Yujian, L., Njuki, S., Yingchun, L.: A comparative study of fine-tuning deep learning models for plant disease identification. Comput. Electron. Agric. 161, 272–279 (2018)CrossRefGoogle Scholar
  23. 23.
    Tushar, A.K., Ashiquzzaman, A., Afrin, A., Islam, M.R.: A novel transfer learning approach upon Hindi, Arabic, and Bangla numerals using convolutional neural networks. In: Computational Vision and Bio Inspired Computing, pp. 972–981 (2018)Google Scholar
  24. 24.
    Wang, X.F., Yu, J., Xie, M.X., Yao, Y.T., Han, J.: Identification and dating of the fountain pen ink entries on documents by ion-pairing high-performance liquid chromatography. Forensic Sci. Int. 180(1), 43–49 (2008)CrossRefGoogle Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2020

Authors and Affiliations

  1. 1.Department of Computer Science and EngineeringIndian Institute of Technology (ISM)DhanbadIndia
  2. 2.Institute for Development and Research in Banking TechnologyHyderabadIndia

Personalised recommendations