Title :
Description method of Japanese manual alphabet using image features
Author :
Ochiai, Katsuyuki ; Kamata, Kazuo
Author_Institution :
Dept. of Inf. Sci., Utsunomiya Univ., Japan
Abstract :
A description is given of a method for handshapes of the Japanese manual alphabet (JMA) based on their contour features. JMA has 46 patterns each of which represents a voiceless sound of the Japanese spoken language. Since 5 fingerspellings have motion, the authors adopt a set of 41 fingerspellings
Keywords :
character recognition; natural languages; Japanese manual alphabet; contour features; fingerspellings; image features; Classification tree analysis; Fingers; Handicapped aids; Image edge detection; Image processing; Image segmentation; Information science; Natural languages; Shape; Wrist;
Conference_Titel :
Systems, Man and Cybernetics, 1989. Conference Proceedings., IEEE International Conference on
Conference_Location :
Cambridge, MA
DOI :
10.1109/ICSMC.1989.71467
