IDENTIFICATION AND CLASSIFICATION OF HISTORICAL KANNADA HANDWRITTEN DOCUMENT IMAGES USING GLCM FEATURES

Main Article Content

Parashuram Bannigidada
Chandrashekar Gudada

Abstract

Our ancestor tried in a many ways to transfer their heritage by inscribing on stone, wooden boards, leafs, metal plates cloths and lastly, on paper, which tends to a long lasting materials. These inscriptions usually contains information related to our civilization, religion, science, astrology, education, etc., and most of these inscriptions are degraded because of its aging, ink bleed-through, folding, etc., due to this degradness, the documents are not possible to read and understand the contents. Hence, it is very much essential to restore by digitising the historical Kannada handwritten documents and also recognise the originality of dynasty it belongs. The main objective of the present study is to digitize, restore and identify the historical Kannada handwritten document images by applying image enhancement techniques; and block wise segmentation method. The identification and classification is done by extracting GLCM features and LDA, K-nearest neighbour (K-NN) and SVM classifiers. In this paper, we have considered historical Kannada handwritten document images of different dynasties based on their age-type; Vijayanagara dynasty (1460 AD), Mysore Wodeyars dynasty (1936 AD), Vijayanagara dynasty (1400 AD) and Hoysala dynasty (1340 AD) for experimentation. The average classification accuracy for different dynasties are; The LDA classifier has yielded an accuracy of 88.2%, K-NN classifier has got 92.3% and SVM classifier has 96.7%, It is observed that, the SVM classifier has got a good classification performance comparatively LDA and K-NN classifiers for Historical Kannada handwritten document images. The results are also compared with manual results obtained by the Epigraphists and language experts, which demonstrate the efficacy and exhaustiveness of the proposed method.

Downloads

Download data is not yet available.

Article Details

Section
Articles
Author Biographies

Parashuram Bannigidada, Rani Channamma University, Belagavi

Department of Computer Science, Assistant Professor

Chandrashekar Gudada, Rani Channamma University, Belagavi

Department of Computer Science, Research Scholar

References

M. G. Manjunath, G. K. Devarajaswamy,â€Kannada Lipi Vikasaâ€, Book published by Jagadhguru Sri Madhvacharya Trust, Sri Raghavendra Swami Matta, Mantralaya.

A.V.Narasimha Murthy - ‘Kannada Lipiya Ugama Mattu Vikasa’, Book published by Kannada Adhyayana Samsthe, Mysore University, Mysore, 1968.

Devarakonda Reddy : Lipiya Huttu Mattu Belavanige — Origin and Evolution of Script, Published by Kannada Pustaka Pradhikara (Kannada Book Authority), Bangalore

Victor Lavrenko, Toni M. Rath and R. Manmath “Holistic Word Recognition for Handwritten Documents†, DIAL '04 Proceedings of the First International Workshop on Document Image Analysis for Libraries (DIAL'04), pp.278-287, 2004.

Joan Andreu Sanchez, Vicent Bosch, Verónica Romero, Katrien Depuydt and Jesse de Does “A Handwritten text recognition for historical documents in the tranScriptorium Project†DATeCH '14 Proceedings of the First International Conference on Digital Access to Textual Cultural Heritage, 2014, pp. 111-117.

Dinesh Dileep “A Feature Extraction technique based on character geometry for character recognitionâ€, The Computing Research Repository Journal, Vol. abs/1202.3884, pp.1- 4, 2012.

Veronica Romero, Nicolas Serrano, Alejandro H. Toselli, Joan Andreu Sanchez and Enrique Vidal “Handwritten Text Recognition for Historical Documents†Proceedings of Langauage Technologies for Digital heritage and Cultural Heritage Workshop, pp. 90-96.

Guanglai Gao, Xiangdong Su, Hongxi Wei and Yeyum Gong “Classical Mongolian Words Recognition in Historical Documentâ€, IEEE, ICDAR, 1520-5363/11, pp.692-697, 2011.

Soumya A. and G. Hemantha Kumar,“Performance Analysis of Random Forests with SVM and KNN in Classification of Ancient Kannada Scripts†International journal of Computers and Technology, Vol. 13 No.9, pp-4907-4921, 2014.

B. Gangamma, K. Srikanta Murthy and Punitha P, “Curvelet Transform based Approch for Prediction of Epigraphical Scripts Eraâ€, IEEE International Conference on Computational Intelligence and Computing Research, 978-4673-1344/12, pp.1-6, 2012.

Soumya A. and G. Hemantha Kumar “SVM Classifier for the Predication of Era of an Epigraphical Script†International journal of Peer to Peer Networks(IJP2P), Vol. 2, No. 2, pp.12-22, 2011.

Soumya A. and G. Hemantha Kumar “Recognition of Historical Records using Gabor and Zonal Featuresâ€, Signal and Image Processing: An International Journal (SIPIJ) Vol. 6, No. 4, pp.57-69, 2015.

H. S. Mohana, Navya K., P. C. Srikanth, G. Shivakumar, “Stone inscripted Kannada Character matching Using SIFTâ€, Proceeding of IRF International Conference, 2014, pp-126-131.

Verma, Brijesh, Blumenstein, Michael, Kulkarni, S.,“ Recent Achievements in Offline Handwriting Recognition Systemâ€, International Conference on Computational Intelligence and Multimedia Applications, 1998.

Parashuram Bannigidad, Chandrashekar Gudada, “Restoration of Degraded Historical Kannada Handwritten Document Images using Image Enhancement Techniquesâ€, International Conference on Soft Computing and Pattern Recognition (SoCPaR 2016), https://doi.org/10.1007/978-3-319-60618-7_49, Springer, Cham, pp 498-508, 2016.

Parashuram Bannigidad, Chandrashekar Gudada, Restoration of Degraded Kannada Handwritten Paper Inscriptions (Hastaprati) using Image Enhancement Techniques, IEEE International Conference on Computer Communication and Informatics (ICCCI -2017), 10.1109/ICCCI.2017.8117697, pp.1-6, 2017.

http://support.echoview.com/WebHelp/Windows_and_Dialog_Boxes/Dialog_Boxes/Variable_properties_dialog_box/Operator_pages/GLCM_Texture_Features.htm

http://www.statsoft.com/Textbook/Support-Vector-Machines#Classification SVM

https://machinelearningmastery.com/k-nearest-neighbors-for-machine-learning

Partha Pratim Roy, Jean-Yves Ramel, Nicolas Ragot, “Word Retrieval in Historical Document Using Character-Primitivesâ€, IEEE International Conference on Document Analysis and Recognition, 2011, pp.678-682.

Andreas Fischer, Emanuel Indermuhle, Volkmar Frinken and Horst Bunke, “HMM-Based Alignment of Inaccurate Transcription for Historical Documentsâ€, IEEE International Conference on Document Analysis and Recognition, 2011, pp.53-57.

Jie Wang and Chew Lim Tan, “Non-rigid Registration and Restoration of Double-sided Historical Manuscripts†IEEE International Conference on Document Analysis and Recognition, 2011, pp.1374-1378.

Jacques Ryckmans, “Inscribed old south Arabian sticks and palm leaf stalks: An Introduction and a Palographical approach†Proceeding of the Seminar for Arabian Studies, Vol.23, 23rd July 1992, pp.127-140.