Modified Convolutional Neural Network of Tamil Character Recognition

  • C. VinotheniEmail author
  • S. Lakshmana Pandian
  • G. Lakshmi
Conference paper
Part of the Lecture Notes in Networks and Systems book series (LNNS, volume 127)


The handwritten Tamil character recognition in offline mode is challenging tasks as there are virtually different people who have different styles of writing the same characters. Deep convolution neural networks are playing a virtual role nowadays in recognizing handwritten character by automatically learning discriminative features from high dimensionality of input data. This work presents a modified convolution neural network \(\left( \text {M-CNN} \right) \) architecture to achieve a faster convergence rate and also to get the highest recognition accuracy. The M-CNN on different aspects along with layers design, activation function, loss function and optimization is discussed. Systematic experiments on isolated handwritten Tamil character dataset collected from various schools by ourselves. For these collected datasets, the proposed system recognized the characters with 97.07%.


Tamil handwritten recognition character Modified convolutional neural network Activation function 


  1. 1.
    Thadchanamoorthy S, Kodikara ND, Premaretne HL, Pal U, Kimura F (2013) Tamil handwritten city name database development and recognition for postal automation. In: IEEE-ICDAR, pp 793–797Google Scholar
  2. 2.
    Feng B, Ren M, Zhang X, Suen C (2014) Automatic recognition of serial numbers in bank notes. Patt Recogn 47:2621–2634CrossRefGoogle Scholar
  3. 3.
    Zhang X-Y, Bengio Y, Liu C-L (2017) Online and offline handwritten chinese character recognition: a comprehensive study and new benchmark. Patt Recogn 61:348–60CrossRefGoogle Scholar
  4. 4.
    Ashay Singh and Ankur Singh Bist (2019) A wide scale survey on handwritten character recognition using machine learning. Int J Comput Sci Eng 7(6):124–134Google Scholar
  5. 5.
    Gu J, Wang Z, Kuen J, Ma L, Shahroudy A, Shuai B, Liu T et al (2018) Recent advances in convolutional neural networks. 77:354–377Google Scholar
  6. 6.
    Raghupathy KB, Chandrasekaran S (2019) Benchmarking on offline Handwritten Tamil Character Recognition using convolutional neural networks. J King Saud Univ Comput Informa SciGoogle Scholar
  7. 7.
    Sureshkumar C, Ravichandran T (2010) Handwritten Tamil character recognition and conversion using neural network. Int J Comput Sci Eng 02(07):2261–2267Google Scholar
  8. 8.
    Vinotheni C, Lakshmana Pandian L (2019) A state of art approaches on handwriting recognition models. In: IEEE fifth international proceedings on science technology engineering and mathematics, vol 1, pp 98–103Google Scholar
  9. 9.
    Yang W, Jin L, Tao D, Xie Z, Feng Z (2016) A new training method to enhance deep convolutional neural networks for large-scale unconstrained handwritten Chinese character recognition. Patt Recogn 58:190–203CrossRefGoogle Scholar
  10. 10.
    Zhang Y, Liang S, Nie S, Liu W, Peng S (2018) Robust offline handwritten character recognition through exploring writer-independent features under the guidance of printed data. Patt Recogn Lett 106:20–26CrossRefGoogle Scholar
  11. 11.
    Li Z, Teng N, Jin M, Lu H (2016) Building efficient CNN architec-ture for offline handwritten Chinese character recognition. Int J Doc Anal Recogn (IJDAR) 21(4):233–240CrossRefGoogle Scholar
  12. 12.
    Boufenar C, Kerboua A, Batouche M (2018) Investigation on deep learning for off-line handwritten Arabic character recognition. Cogn Syst Res 50:180–195CrossRefGoogle Scholar
  13. 13.
    El-Sawy A, Loey M, Hazem EB (2017) Arabic handwritten characters recognition using convolutional neural network (2017) . WSEAS Trans Comput Res 5:11–19Google Scholar
  14. 14.
    Roy S, Das N, Kundu M, Nasipuri M (2017) Handwritten isolated Bangla compound character recognition: a new benchmark using a novel deep learning approach. Patt Recogn Lett 90:15–21CrossRefGoogle Scholar
  15. 15.
    Jangid M, Srivastava S (2018) Handwritten Devanagari character recognition using layer-wise training of deep convolutional neural networks and adaptive gradient methods. J Imag 4(2):1–14Google Scholar
  16. 16.
    Raymond P, Petroski F, Such SP, Brockler F, Singh V, Hutkowski P (2018) Intelligent character recognition using fully convolutional neural net-works. Patt Recogn 88:604–613Google Scholar
  17. 17.
    Davit S, Magda T, Levan S, Irakli K, Shota A, Sandro J (2018) On Georgian handwritten character recognition. IFAC-PapersOnLine 51(30):161–165Google Scholar
  18. 18.
    Deepa M, Deepa R, Meena R, Nandhini R (2019) Tamil handwritten text recognition using convolutional neural networks. Int J Eng Sci Comput 9(3):20986–20988Google Scholar
  19. 19.
    Vijayaraghavan, Prashanth, and Misha Sra: Handwritten Tamil recognition using a convolutional neural network. MIT Media Lab, (2014)Google Scholar
  20. 20.
    Niu X-X, Suen CY (2012) A novel hybrid CNN–SVM classifier for recognizing handwritten digits. Patt Recogn 45(4):1318–1325Google Scholar
  21. 21.
    Sahare P, Dhok SB (2018) Multilingual character segmentation and recognition schemes for Indian document images. IEEE Access 6:10603–10617CrossRefGoogle Scholar
  22. 22.
    Pragathi MA, Priyadarshini K, Saveetha S, Shavar Banu A, Mohammed Aarif KO (2019) Handwritten Tamil Character Recognition using deep learning. In: IEEE international proceedings on vision towards emerging trends in communication and networking (ViTECoN), pp 1–5Google Scholar
  23. 23.
    Abadi M, Agarwal A, Barham P, Brevdo E, Chen Z, Citro C, Corrado GS et al (2016) Tensorflow: large-scale machine learning on heterogeneous distributed systems. arXiv preprint arXiv:1603.04467

Copyright information

© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021

Authors and Affiliations

  • C. Vinotheni
    • 1
    Email author
  • S. Lakshmana Pandian
    • 1
  • G. Lakshmi
    • 1
  1. 1.Pondicherry Engineering CollegePondicherryIndia

Personalised recommendations