An Implementation of KSSL Recognizer for HCI Based on Post Wearable PC and Wireless Networks

Abstract

A traditional study on recognition and representation of sign language based on the desktop PC have general restrictions conditionality in space and limitation of motion according to the technology of wire communication, and then the vision technology-based traditional sign langage recognition systems, recognition performance has the possibility of change and diversity according to the performance of the camera, change of background (colors) and illumination condition. To overcome these restrictions and problems, we implement embedded-sentential the Korean Standard Sign Language (hereinafter, KSSL) recognizer based on the post wearable PC for mobile and ubiquitous computing. The KSSL is a dialog system and interactive elements in the South Korea deaf communities. The advantages of our approach are as follows: 1) it improves efficiency of the KSSL input module according to the technology of wireless communication and user need not be constrained by the limitations of a particular interaction mode at any given moment, 2) it recognizes and represents continuous the KSSL of users with flexibility in real time and can offer to user a wider range of personalized and differentiated information more effectively, and 3) it is possible more effective and free interchange of ideas and information between deaf person and hearing person (the public). Experimental result shows an average recognition rate of 93.3% for the KSSL sentences and 96.9% for the KSSL words with significant, dynamic and continuous the KSSL.