Call for Paper - October 2019 Edition
IJCA solicits original research papers for the October 2019 Edition. Last date of manuscript submission is September 20, 2019. Read More

Handwritten Bangla Character Recognition using Inception Convolutional Neural Network

Print
PDF
International Journal of Computer Applications
Foundation of Computer Science (FCS), NY, USA
Year of Publication: 2018
Authors:
Md. Adnan Taufique, Farhana Rahman, Md. Imrul Kayes Pranta, Nasib AL Zahid, Syeda Shabnam Hasan
10.5120/ijca2018917850

Md. Adnan Taufique, Farhana Rahman, Md. Imrul Kayes Pranta, Nasib AL Zahid and Syeda Shabnam Hasan. Handwritten Bangla Character Recognition using Inception Convolutional Neural Network. International Journal of Computer Applications 181(17):48-59, September 2018. BibTeX

@article{10.5120/ijca2018917850,
	author = {Md. Adnan Taufique and Farhana Rahman and Md. Imrul Kayes Pranta and Nasib AL Zahid and Syeda Shabnam Hasan},
	title = {Handwritten Bangla Character Recognition using Inception Convolutional Neural Network},
	journal = {International Journal of Computer Applications},
	issue_date = {September 2018},
	volume = {181},
	number = {17},
	month = {Sep},
	year = {2018},
	issn = {0975-8887},
	pages = {48-59},
	numpages = {12},
	url = {http://www.ijcaonline.org/archives/volume181/number17/29918-2018917850},
	doi = {10.5120/ijca2018917850},
	publisher = {Foundation of Computer Science (FCS), NY, USA},
	address = {New York, USA}
}

Abstract

With the advancement of modern technology the necessity of pattern recognition has increased a lot. Character recognition it's part of pattern recognition. In last few decades there has been some researches on optical character recognition(OCR) for so many languages like - Roman, Japanese, African, Chinese, English and some researches of Indian language like -Tamil, Devanagari, Telugu, Gujratietc and so many other languages. There are very few works on handwritten Bangla character recognition. As it is tough to understand like Bangla language because of different people handwritten varies in fervidity or formation, stripe and angle. For this it's so much challenging to work in this field. In some researches SVM, MLP, ANN, HMM, HLP & CNN has been used for handwritten Bangla character recognition. In this paper an attempt is made to recognize handwritten Bangla character using Convolutional Neural Network along with the method of inception module without feature extraction. The feature extraction occurs during the training phase rather than the dataset preprocessing phase. As CNN can't take input data that varying in shape ,so had to rescaled the dataset images at fixed different size. In total final dataset contains 100000 images of dimension 28x28. 85000 images is used for training and 3000 images is used for testing. After analyzing the results a conclusion is derived on the proposed work and stated the future goals and plans to achieve highest success and accuracy rate.

References

  1. Williams, Kyle, Hussein Suleman, and Jorgina K. do R Paihama. "A comparison of machine learning techniques for handwritten." Proceedings of the South African Institute for Computer Scientists and Information Technologists Conference.ACM, 2013.
  2. Vijayaraghavan, Prashanth, and Misha Sra. "Handwritten Tamil Recognition using a Convolutional Neural Network",2015.
  3. Shanthi, N. and Duraiswamy, K., 2010. A novel SVM-based handwritten Tamil character recognition system. Pattern Analysis and Applications, 13(2), pp.173180.
  4. Kumar, Parveen, Nitin Sharma, and ArunRana. "Handwritten Character Recognition using Different Kernel based SVM Classifier and MLP Neural Network (A COMPARISON)." International Journal of Computer Applications53.11 (2012).
  5. Das, Nibaran, et al. "Handwritten Bangla character recognition using a soft computing paradigm embedded in two pass approach." Pattern Recognition48.6 (2015): 2054-2071.
  6. Rahman, Md Mahbubar, et al. "Bangla handwritten character recognition using convolutional neural network." International Journal of Image, Graphics and Signal Processing (IJIGSP) 7.8 (2015): 42.
  7. Bhowmik, Tapan Kumar, et al. "SVM-based hierarchical architectures for handwritten Bangla character recognition." International Journal on Document Analysis and Recognition (IJDAR) 12.2 (2009): 97-108.
  8. Rahman, Ahmad Fuad Rezaur, R. Rahman, and Michael C. Fairhurst. "Recognition of handwritten Bengali characters: a novel multistage approach." Pattern Recognition 35, no. 5 (2002): 997-1006.
  9. Khan, Haider Adnan, Abdullah Al Helal, and Khawza I. Ahmed. "Handwritten bangla digit recognition using sparse representation classifier." Informatics, Electronics & Vision (ICIEV), 2014 International Conference on. IEEE, 2014.
  10. Bhowmik, T.K., Bhattacharya, U. and Parui, S.K., 2004, November. Recognition of Bangla handwritten characters using an MLP classifier based on stroke features. In International Conference on Neural Information Processing(pp. 814-819).Springer, Berlin, Heidelberg.
  11. Alom, MdZahangir, et al. "Handwritten Bangla Digit Recognition Using Deep Learning." arXiv preprint arXiv:1705.02680 (2017).
  12. Szegedy, Christian, et al. "Going deeper with convolutions." Proceedings of the IEEE conference on computer vision and pattern recognition. 2015.
  13. Scherer, Dominik, Andreas Müller, and Sven Behnke. "Evaluation of pooling operations in convolutional architectures for object recognition." Artificial Neural Networks–ICANN 2010 (2010): 92-101.
  14. Discussion about Convolutional Neural Network http://ufldl.stanford.edu/tutorial/supervised/ConvolutionalNeuralNetwork (LAT*-16 july,2017,11:06pm) .
  15. Stanford cs231n lectures on CNN http://cs231n.github.io /convolutional-networks (LAT*-15 april,2018,12:35am).
  16. A Beginner's Guide To Understanding Convolutional Neural Network https://adeshpande3.github.io/ adeshpande3.github.io/A-Beginner%27s-Guide-ToUnderstanding-Convolutional-Neural-Networks (LAT*-20 june,2017,9:21pm).
  17. Discussion about input dimension of CNNs https://www.quora.com/Does-input-dimension-resolution-affect-the-performance-of-Convolutional-Neural-Networks. (LAT*-30 april,2018,12:00am).
  18. Coursera DNN specialization course lectures https://www.coursera.org/learn/neural-networks-deeplearning/ lecture/A0tBd/ gradient-descent(LAT*-28 January,2018,10:02am).
  19. Selected dataset repository on google code http://code.google.com/p/cmaterdb(LAT*-18 june,2017,
  20. 6:42 pm).
  21. Discussion the characterization of epoch and iteration https://stackoverflow.com/questions/4752626/epoch-vs-iteration-when-training neural-networks(LAT*-28 November,2017,10:s02am).
  22. Baldi, P. and Sadowski, P.J., 2013. Understanding dropout. In Advances in neural information processing systems (pp. 2814-2822).
  23. Kanan, Christopher, and Garrison W. Cottrell. "Color-to-grayscale: does the method matter in image recognition?." PloS one 7.1 (2012): e29740.
  24. Article explaining CNNs on a deeper level https://ujjwalkarn.me/2016/08/11/intuitive-explanation-convnets(LAT*-23 july, 2017,8:50pm) .
  25. Article explaining inception modules https://hacktilldawn.com/2016/09/25/inception-modules-explained-and-implemented(LAT*-18april,2018,10:15
  26. pm).
  27. Discussion on the process of calculating the number of parameters of a convolutional neural network https://stackoverflow.com/questions/42786717/how-to-calculate-the-number-of-parameters-for-convolutiona-neural-neural-network/42787467(LAT*-25april,2018, 6:45pm).
  28. Coursera DNN lecture on activation functions https://www.coursera.org/learn/neural-networks-deep-learning/lecture/ OASKH/ why-do-you-need-non-linear-activationfunctions(LAT*-28january,2018,10:03am).

Keywords

Handwritten Bangla character, Shallow convonet, CNN, Inception, Data Normalization