Abstract
In order to rapidly build an automatic and precise system for image recognition and categorization, deep learning is a vital technology. Handwritten character classification also gaining more attention due to its major contribution in automation and specially to develop applications for helping visually impaired people. Here, the proposed work highlighting on fine-tuning approach and analysis of state-of-the-art Deep Convolutional Neural Network (DCNN) designed for Devanagari Handwritten characters classification. A new Devanagari handwritten characters dataset is generated which is publicly available. Datasets consist of total 5800 isolated images of 58 unique character classes: 12 vowels, 36 consonants and 10 numerals. In addition to this database, a two-stage VGG16 deep learning model is implemented to recognize those characters using two advanced adaptive gradient methods. A two-stage approach of deep learning is developed to enhance overall success of the proposed Devanagari Handwritten Character Recognition System (DHCRS). The first model achieves 94.84% testing accuracy with training loss of 0.18 on new dataset. Moreover, the second fine-tuned model requires very fewer trainable parameters and notably less training time to achieve state-of-the-art performance on a very small dataset. It achieves 96.55% testing accuracy with training loss of 0.12. We also tested the proposed model on four different benchmark datasets of isolated characters as well as digits of Indic scripts. For all the datasets tested, we achieved the promising results.
Similar content being viewed by others
Dataset accessibility
Our newly created Handwritten Devanagari character dataset is publically available at https://www.kaggle.com/shalakadeore/handwritten-marathi-devanagari-characters.
References
Bishop C 2006 Pattern Recognition and Machine Learning. Springer, Berlin
Basu S, Das N, Sarkar R, Kundu M, Nasipuri M and Basu D K 2010 A novel framework for automatic sorting of postal documents with multi-script address blocks. Pattern Recogn. 43(10): 3507–3521
Roy K, Vajda S, Pal U, Chaudhuri B B and Belaid A 2005 A system for Indian postal automation. In: Proceedings of the Eighth International Conference on Document Analysis and Recognition, IEEE Computer Society, USA, pp. 1060–1064
Pham T A, Le H H and Do N T 2015 Offline handwritten signature verification using local and global features. Ann. Math. Artif. Intell. 75: 231–247
Palacios R and Gupta A 2008 A system for processing handwritten bank checks automatically. Image Vis. Comput. 26: 1297–1313
Romero V, Serrano N, Toselli A H, Sanchez J A and Vidal E 2011 Handwritten Text Recognition for Historical Documents. In: Proceedings of the workshop on Language Technologies for Digital Humanities and Cultural Heritage, Hissar, Bulgaria, pp. 90–96
Banerjee P, Bhattacharya U and Chaudhuri B B 2014 Automatic Detection of Handwritten Texts from Video Frames of Lectures. In: Proceedings of 14th International Conference on Frontiers in Handwriting Recognition, Heraklion, pp. 627–632
Liu C L, Nakashima K, Sako H and Fujisawa H 2003 Handwritten digit recognition: benchmarking of state-of-the-art techniques. Pattern Recogn. 36(10): 2271–2285
Singh P K, Sarkar R and Nasipuri M 2015 Offline Script Identification from multilingual Indic-script documents: a state-of-the-art. Comput. Sci. Rev. 15-16: 1–28
Pal U and Chaudhuri B B 2004 Indian script character recognition: A survey. Pattern Recogn. 37(9): 1887–1899
Deore S P, Pravin A 2017 Ensembling: Model of histogram of oriented gradient based handwritten Devanagari character recognition system. Traitement du Signal. 34(1-2): 7–20
Ruck D W, Rogers S K and Kabrisky M 1990 Feature selection using a multilayer perceptron. J. Neural Netw. Comput. 2(2): 40–48
Yang J, Shen K, Ong C and Li X 2009 Feature selection for mlp neural network: The use of random permutation of probabilistic outputs. IEEE Trans. Neural Netw. 20(12): 1911–1922
Lee H, Grosse R, Ranganath R and Ng A Y 2009 Convolutional deep belief networks for scalable unsupervised learning of hierarchical representations. In: Proceedings of the 26th Annual International Conference on Machine Learning. ACM, pp. 609–616
Liu W, Wang Z, Liu X, Zeng N, Liu Y and Alsaadi F E 2017 A survey of deep neural network architectures and their applications. Neurocomputing 234: 11–26
Niu X and Suen C Y 2012 A novel hybrid CNN–SVM classifier for recognizing handwritten digits. Pattern Recogn. 45: 1318–1325
Lauer F, Suen C Y and Bloch G 2007 A trainable feature extractor for handwritten digit recognition. Pattern Recogn. 40: 1816–1824
Alom M Z, Sidike P, Taha T M and Asari V K 2018 Handwritten Bangla character recognition using the state-of-the-art deep convolutional neural networks. Comput. Intell. Neurosci. 2018: 1–13
Younis K 2017 Arabic handwritten character recognition based on deep convolutional neural networks. Jordan. J. Comput. Inf. Technol. 3(3):186-200
Ukil S, Ghosh S, Obaidullah S M, Santosh K C, Roy K and Das N 2019 Improved word-level handwritten Indic script identification by integrating small convolutional neural networks. Neural Comput. Appl. 32: 2829–2844
Jangid M and Srivastava S 2018 Handwritten Devanagari character recognition using layer-wise training of deep convolutional neural networks and adaptive gradient methods. J. Imaging. 4: 1–14
Sarkhel R, Das N, Das A, Kundu M and Nasipuri M 2017 A multi-scale deep quad tree based feature extraction method for the recognition of isolated handwritten characters of popular indic scripts. Pattern Recogn. 71: 78–93
Gupta A, Sarkhel R, Das N and Kundu M 2019 Multiobjective optimization for recognition of isolated handwritten Indic scripts. Pattern Recogn. Lett. 128: 318-325
Acharya S, Pant A and Gyawali P 2015 Deep Learning Based Large Scale Handwritten Devanagari Character Recognition. In: Proceedings of the 9th International Conference on Software, Knowledge, Information Management and Applications. Kathmandu, pp. 1–6
Aneja N and Aneja S 2019 Transfer Learning using CNN for Handwritten Devanagari Character Recognition. In: Proceeding of the 1st International Conference on Advances in Information Technology. Chikmagalur, India, pp. 293–296.
Guha R, Das N, Kundu M, Nasipuri M and Santosh K C 2019 DevNet: An efficient CNN architecture for handwritten Devanagari character recognition. Int. J. Pattern Recogn. Artif. Intell.
Rahman Md M, Akhand M A H, Islam S, Shill P C and Rahman M M H 2015 Bangla handwritten character recognition using convolutional neural network. Int. J. Image Graph. Signal Process. 7: 42–49
Bhattacharya U, Shridhar M and Parui S 2006 On recognition of handwritten Bangla characters. In: Computer Vision, Graphics and Image Processing, Lecture Notes in Computer Science. Springer, Berlin, Heidelberg, 4338, pp. 817–828
Purkaystha B, Datta T and Islam M 2017 Bengali handwritten character recognition using deep convolutional neural network. In: Proceeding of 20th International Conference of Computer and Information Technology. Dhaka, pp. 1–5
Hassan T and Khan H 2015 Handwritten Bangla numeral recognition using Local Binary Pattern. In: Proceeding of International Conference on Electrical Engineering and Information Communication Technology. Dhaka, pp. 1–4
Saha C, Faisal R and Mostafijur R 2018 Bangla Handwritten Character Recognition Using Local Binary Pattern and Its Variants. In: Proceeding of International Conference on Innovations in Science, Engineering and Technology. Chittagong, Bangladesh, pp. 236–241
Kobetski M and Sullivan J 2013 Apprenticeship learning: transfer of knowledge via dataset augmentation. In: Proceeding of the Scandinavian Conference on Image Analysis. Lecture Notes in Computer Science. Springer, Berlin, Heidelberg, 7944, pp. 432–443
Perez L and Wang J 2017 The effectiveness of data augmentation in image classification using deep learning. In: arXiv preprint arXiv:1712.04621
Russakovsky O, Deng J and Su H 2015 Imagenet large scale visual recognition challenge. Int. J. Comput. Vis. 115: 211–252
Simonyan K and Zisserman A 2015 Very Deep Convolutional Networks for Large-Scale Image Recognition. In: arXiv:1409.1556v6
Balci B, Saadati D and Shiferaw D 2017 Handwritten Text Recognition Using Deep Learning. In: CS231n: Convolutional Neural Networks for Visual Recognition, Stanford University. Course Project Report
Hinton G coursera course - lecture 6e 2012. http://www.cs.toronto.edu/tijmen/csc321/slides/lecture_slides_lec6.pdf
Kingma D P and Lei Ba J 2015 Adam: a Method for Stochastic Optimization. In: Proceeding of the International Conference on Learning Representations. pp. 1–13
Nair V and Hinton G E 2010 Rectified linear units improve restricted Boltzmann machines. In: Proceedings of the 27th International Conference on Machine Learning. Haifa, Israel, pp. 807–814
Srivastava N, Hinton G, Krizhevsky A, Sutskever I and Salakhutdinov R 2014 Dropout: A simple way to prevent neural networks from overfitting. J. Mach. Learn. Res. 15(1): 1929–1958
Passricha V and Aggarwal R K 2019 A comparative analysis of pooling strategies for convolutional neural network based Hindi ASR. J. Ambient Intell. Humaniz. Comput. 11: 675–691
Too E C, Yujian L, Njuki S and Yingchun L 2019 A comparative study of fine-tuning deep learning models for plant disease identification. Comput. Electron. Agric. 161: 272–279
Ostler N 2005 Empires of the Word: A Language History of the World. HarperCollins, New York
Emeneau M 1956 India as a Lingustic Area. Language (Baltim). 32: 3–16
Sarkhel R, Das N, Basu S, Kundu M and Nasipuri M 2012 CMATERdb1: a database of unconstrained handwritten Bangla and Bangla – English mixed script document image. Int. J. Doc. Anal. Recogn. 15: 71–83
Das N, Reddy J M, Sarkar R, Basu S, Kundu M, Nasipuri M and Basu D K 2012 A statistical–topological feature combination for recognition of handwritten numerals. Appl. Soft Comput. 12: 2486–2495
Sarkar A, Singh K and Mukerjee A 2012 Handwritten Hindi Numerals Recognition System. Webpage: https://www.cse.iitk.ac.in/users/cs365/2012/submissions/aksarkar/cs365, CS365 project report
Roy A, Das N, Sarkar R, Basu S and Kundu M 2014 An Axiomatic Fuzzy Set Theory Based Feature Selection Methodology for Handwritten Numeral Recognition. In: ICT and Critical Infrastructure: Proceedings of the 48th Annual Convention of Computer Society of India- Vol I. Advances in Intelligent Systems and Computing. 248, pp. 133–140
Vijaya Kumar R and Babu U 2019 Handwritten Hindi character recognition using deep learning techniques. Int. J. Comput. Sci. Eng. 7(2): 1–7
Saha P and Jaiswal A 2020 Handwriting recognition using active contour. In: Proceeding of the Artificial Intelligence and Evolutionary Computations in Engineering Systems. Advances in Intelligent Systems and Computing. Springer, Singapore, 1056, pp. 505–514
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Deore, S.P., Pravin, A. Devanagari Handwritten Character Recognition using fine-tuned Deep Convolutional Neural Network on trivial dataset. Sādhanā 45, 243 (2020). https://doi.org/10.1007/s12046-020-01484-1
Received:
Revised:
Accepted:
Published:
DOI: https://doi.org/10.1007/s12046-020-01484-1