Abstract:
Online Sinhala handwritten character recognition is a fairly under-developed, albeit well
researched area in the local natural language initiative. Recent attention towards this area
produced an application named PanhinDA, a Sinhala Handwritten Character Recognition
Tool for PDAs from the University of Moratuwa, Dept. of Computer Science and
Engineering. However, due to the obvious lack of high processing capability in a PDA,
development of this tool was stunted, using just one method of recognition. This resulted
in poor recognition capabilities. Nevertheless, the development of such a tool for
recognising Sinhala handwritten characters is essential, playing a vital role when the
nation moves towards e-governance. Such a tool could also be used for academic
purposes. This has prompted further research in the area, with the aim of incorporating
multiple (alternative) techniques simultaneously to ensure higher recognition. Sinhala
consists of 18 vowels, 41 consonants and 17 character associative symbols, and unlike
English, poses a unique array of difficulties, since the letters are of a curved form, as
opposed to straight lines.