Identification and classification of historical Kannada handwritten document images using LBP features
by Parashuram Bannigidad; Chandrashekar Gudada
International Journal of Intelligent Systems Design and Computing (IJISDC), Vol. 2, No. 2, 2018

Abstract: Digitisation process is very much essential for historical handwritten documents due to degraded qualities in the manuscript such as ink bleed, arbitrary geometric distortions, storage conditions and age. In this work, the age-type identification and classification of historical Kannada handwritten document images is done by applying text-block wise segmentation method, extracting the LBP features and using LDA, K-NN and SVM classifiers. The purpose of the present work is to identify the document script of the dynasties, whether it belongs to Hoysala dynasty or Vijayanagara dynasty or Mysore Wodeyar's dynasty? The average classification accuracy of the proposed method with LDA classifier is 94.6%, whereas K-NN classifier yielded 98.3% and SVM classifier yielded 99.3%. As per the results obtained from the experimentation, it is proved that the SVM classifier has got good classification ability comparatively with respect to LDA and K-NN classifiers for historical Kannada handwritten document images of all age-type scripts.

Online publication date: Mon, 26-Nov-2018

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Intelligent Systems Design and Computing (IJISDC):
Login with your Inderscience username and password:

    Username:        Password:         

Forgotten your password?


Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com