Advertisement

Analyzing the Effect of Regularization and Augmentation in Deep Neural Network Model with Handwritten Digit Classifier Dataset

  • P. Madhan RajEmail author
  • B. Arun Kumar
  • G. Bharath
  • S. Murugavalli
Conference paper
Part of the Lecture Notes on Data Engineering and Communications Technologies book series (LNDECT, volume 35)

Abstract

A lot of research has been carried out in the field of Handwritten Digit Recognition in recent years. It has its application in areas like bank check processing, signature verification, etc. where very high level of accuracy is a required and even a small mistake would lead to a great loss of money and time. I propose a model in my system with a accuracy of 99.6% using Deep Learning neural networks assisted by Data Augmentation and Regularization.

Keywords

Handwritten digit recognition Neural network Deep learning 

References

  1. 1.
    Ashiquzzaman, A., Tushar, A.K.: Handwritten Arabic numeral recognition using deep learning neural networks. In: 2017 IEEE International Conference on Imaging, Vision and Pattern Recognition (icIVPR), Dhaka, pp. 1–4 (2017). Md Shopon, Nabeel Mohammed and Md Anowarul Abedin (2017)Google Scholar
  2. 2.
    Shopon, M., et.al.: Image augmentation by blocky artifact in deep convolutional neural network for handwritten digit recognition. In: 2017 IEEE International Conference on Imaging, Vision and Pattern Recognition (icIVPR), Dhaka, pp. 1–6 (2017)Google Scholar
  3. 3.
    Li, X., et.al.: FPGA accelerates deep residual learning for image recognition. In: 2017 IEEE 2nd Information Technology, Networking, Electronic and Automation Control Conference (ITNEC), Chengdu, pp. 837–840 (2017)Google Scholar
  4. 4.
    Shopon, M., Mohammed, N., Abedin, M.A.: Bangla handwritten digit recognition using auto encoder and deep convolutional neural network. In: 2016 International Workshop on Computational Intelligence (IWCI), Dhaka, pp. 64–68 (2016)Google Scholar
  5. 5.
    Saabni, R.: Recognizing handwritten single digits and digit strings using deep architecture of neural networks. In: 2016 Third International Conference on Artificial Intelligence and Pattern Recognition (AIPR), Lodz, pp. 1–6 (2016)Google Scholar
  6. 6.
    Kiani, K., Korayem, E.M.: Classification of Persian handwritten digits using spiking neural networks. In: 2015 2nd International Conference on Knowledge-Based Engineering and Innovation (KBEI), Tehran, pp. 1113–1116 (2015)Google Scholar
  7. 7.
    Agapitos, A., et. al.: Deep evolution of image representations for handwritten digit recognition. In: 2015 IEEE Congress on Evolutionary Computation (CEC), Sendai, pp. 2452–2459 (2015)Google Scholar
  8. 8.
    Srivatsa, N., et al.: Dropout: a simple way to prevent neural networks from over fitting. J. Mach. Learn. Res. 15(2014), 1929–1958 (2014)MathSciNetGoogle Scholar
  9. 9.
    Zhang, S., et.al.: Learning high-level features by deep Boltzmann machines for handwriting digits recognition. In: Proceedings of 2nd International Conference on Information Technology and Electronic Commerce, Dalian, pp. 243–246 (2014)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  • P. Madhan Raj
    • 1
    Email author
  • B. Arun Kumar
    • 1
  • G. Bharath
    • 1
  • S. Murugavalli
    • 1
  1. 1.Computer Science and EngineeringPanimalar Engineering CollegeChennaiIndia

Personalised recommendations