Recognition based classification of Gurmukhi manuscripts
Simpel Rani, Gurpreet Singh Lehal · 2016
Most of the commercial optical character recognition (OCR) software are designed to recognize the documents which are used in business. The accuracy of an OCR software is high to recognize the documents which are finely printed and fine quality written documents. However, this accuracy decreases drastically when dealing with the documents found in most of the libraries, especially manuscripts. With many collections, the limitations of commercial optical character recognition software become apparent. The manuscripts in any script contain much degradation which may decrease the performance of OCR software. The manuscripts may contain degradation of many types. In this paper, we have made a recognition based classification of the Gurmukhi manuscripts into various categories. The different styles of manuscripts, the different kinds of degradation associated with each style and recognition problems in ancient manuscripts of Gurumukhi script have been identified. Along with identifying the different styles of manuscripts, the problems associated with each kind of manuscript have been discussed and some possible solutions have also been proposed. This paper is extremely useful for researchers engaged in recognizing the manuscript documents in any script, because same kinds of problems can be found in manuscript document of any script of the world.