Paper Title
Stone In-Scripted Kannada Characters Matching Using Sift

Every nation is recognized by its tradition and cultural heritage. India is a country with its rich and glorious history. The Historical evidences are written documents in the form of various inscriptions and manuscripts. A great deal of effort is being made to prevent their degradation from globalization and natural calamities. A lot of research is also done in finding methods to digitize these documents so that they can immortalize. Saving those signs of history and cultural heritage are most crucial for not only India but also for every other nation. This paper is based on the method for matching stone inscripted Kannada characters using the Kannada stone inscriptions images by Scale Invariant Feature Transform Algorithm. This approach of matching characters can robustly match the characters among the group of characters with various sizes and shapes, achieving near real-time performance. The features of a Kannada character is highly distinctive, in the sense that a single feature can be correctly matched with high probability against another Kannada character features. Final verification of matching is achieved by finding the edges and Euclidian distance between all the characters in the images of stone inscriptions. Experimental results show that robust character matching can be achieved with less computation time. This paper describes the digitization of stone inscriptions available in and around Hassan, Karnataka state, India for documentation and matching of characters in the stone inscriptions. Further, the matched images of stone inscriptions are restored and processed for the purpose of archaeological studies. The authors achieve the matching accuracy of up to 89.5% with the computational time of not more than 15 seconds for complete system execution. This proposed work helps in the preservation of cultural heritage of our nation.