| Summary: | 碩士 === 淡江大學 === 機械與機電工程學系碩士班 === 100 === The objective of this thesis is to improve input efficiency of communication device for visually impaired people with multiple disabilities using initial consonant to simplify ZhuYin input method. At first, the images of head shaking are captured through an image device in the study by the Haar facial feature detection and tracking systems. The moving position of the nose center is computed to identify the directions of head shaking, and then using initial consonant to simplify ZhuYin input method with the voice communication and text output are achieved by the combined text-to-speech editor, and accomplished the communication with the outside world. The input method has a problem that the simplified inputs often generate many corresponding words. This study creates the one-to-one thesaurus using all the phonetic symbols of the last word to solve the problem that the visually impaired people with multiple disabilities can not easily choose the different words by shaking head to operate the communication device.
In this study, the newly accomplished communication device combined the Haar facial feature detection and tracking systems with initial consonant to simplify ZhuYin input method had been tested. Besides, the input efficiency of the new communication device was compared with the input efficiency of the communication device that formerly combined the Haar facial feature detection and tracking systems with ordinary phonetic input method. The result showed that the input efficiency of the new communication device increased by 53.61%. The new communication device effectively improves the input efficiency.
The results of this study can reduce the shaking head times to effectively enhance the efficiency of communication device for visually impaired people with multiple disabilities, and can also create the quality assured accessibility environment of learning and living.
|
|---|
