• PARASHURAM BANNIGIDAD
• CHANDRASHEKAR GUDADA

The inscriptions or Epigraphical manuscripts where composed on various material, for example, walls on the caves, stone carving, palm leaf, metal plates and paper are the resources and cultural heritage of our country; our focus is to reproduce the cultural significance of the Kannada Language and its traditional writing through the historical manuscripts. The majority of the assets are in the degraded state, the degraded manuscripts are affected by many factors like, weather condition, ink bleed through and quality of the writing material. The offline handwritten text recognition is one of the most challenging tasks in document image analysis. In the present era of digital, it’s our fundamental duty to protect the resources of our Indian culture and heritage by digitizing the manuscripts which are losing its originality and status. In this paper, we are trying to identify and recognize the historical Kannada handwritten scripts of various dynasties; namely, Vijayanagara dynasty (1460 AD), Mysore Wadiyar dynasty (1936 AD), Vijayanagara dynasty (1400 AD) and Hoysala dynasty (1340 AD) by using seam carving line segmentation method by extracting LBP features. For recognition and classification; the LDA, K-NN and SVM classifiers are used. The average classification accuracy for different dynasties is computed. The computed results are 94.2% for LDA, 94.9% for K-NN and 96.4% for SVM is achieved, based on the experimented result the SVM classifier is yielded the higher classification accuracy comparatively LDA and K-NN for historical Kannada handwritten scripts. The experimental outcomes are verified by language experts and Epigraphists

Document image analysis, Historical documents, Handwritten script, Seam carving, Line segmentation, Kannada, LBP, LDA, K-NN, SVM

"Digitization and Recognition of Historical Kannada Handwritten Manuscripts using Text Line Segmentation with LBP Features", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 4, page no.657-664, April-2019, Available :http://www.jetir.org/papers/JETIR1904997.pdf

"Digitization and Recognition of Historical Kannada Handwritten Manuscripts using Text Line Segmentation with LBP Features", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 4, page no. pp657-664, April-2019, Available at : http://www.jetir.org/papers/JETIR1904997.pdf