Formulation of an image processing technique for improving Sign2 performance
Kanthi Sarella · RIT Scholar Works (Rochester Institute of Technology) · 2006
This paper states and discusses an image processing technique for improving the performance of Phase I of the Sign2 Project. The Sign2 Project is a complete technological approach to the translation of ASL to digital audio and/or text. Considerable research had been carried out in the Sign2 Project and significant progress had been achieved. Tangible results were derived from the research but there had been unintended inconsistencies in the results where accuracy was concerned. This paper discusses the possible reasons for failure of the Sign2 Project to achieve the intended accuracy and proposes potential methods to ameliorate the approach to achieve better results.