Language Technology Research Laboratory carries out various research activities on issues related to Sinhala in the context of computational linguistics and Natural Language Processing. Some of them are done in collaboration with outsiders and some personally depending on the interests of members.
Some of the main topics are:
|
.....................................................................................................................................
|
Online Handwriting Character Recognition
The research is focused on developing an Online Handwriting Character Recognition (OLCR) system for Sinhala. Initially, a character-based recognizer will be implemented. Then, this solution will be extended to recognize words.
Due to the proven better performance of template-based elastic matching (EM) algorithms for the Indic languages, it is expected to attempt and investigate the performance of EM algorithms for the recognition of Sinhala handwriting characters. At the initial stages, simple application of k-nearest neighbor algorithm with Euclidian distance is to be investigated following the work that has been done at LTRL when developing the Optical Character Recognizer. The study will then be extended to examine the application of complex and widely used Dynamic Programming (DP) and Dynamic Time Warping (DTW) techniques for Sinhala OLCR.
It is also expected to collect a database of online handwritten Sinhala characters/words from this project.
Project Duration: 3 yrs
Researchers: Asanka Wasala
Supervisor: Dr. Ruvan Weerasinghe
|
