DocumentCode
2353241
Title
A Research on the Stroke-Segment-Mesh (SSM) Glyph Depiction Method of Chinese Character
Author
Lin, Min ; Song, Rou ; Ge, Shi-Li
Author_Institution
Coll. of Comput. Sci. & Technol., Beijing Univ. of Technol., Beijing
fYear
2008
fDate
23-25 July 2008
Firstpage
269
Lastpage
278
Abstract
There are defects in the current depiction method of computerized Chinese character glyph, especially on the aspect of feature selection and computation for comparing glyphs, which cause many problems, such as the input of wrongly written characters, variant forms of characters in ancient literatures, and combined-characters and automatic glyph comparing. This research offers an application-oriented depiction method of Chinese character glyph, which possesses the characteristics of suitable granular degree, ambiguity-free, and standardized basic units and which can be applied in the depiction of any possible different glyph skeletons (including wrongly written characters, variant form of a Chinese character, and combined-characters). This research also offers the classifications of basic stokes, compound stokes, algorithms of automatic stoke extraction and structural relation computation based on stroke-segments-mesh depiction of glyphs. The experiment indicates that this method can be applied in the descriptive input of any glyph and computation for comparing either the whole or part of glyphs.
Keywords
character recognition; natural languages; Chinese character glyph; application-oriented depiction method; automatic stoke extraction; stroke-segment-mesh glyph depiction; Character recognition; Computer science; Cultural differences; Education; Educational institutions; Handwriting recognition; Humans; Information technology; Natural languages; Skeleton; Chinese Character Glyph; Formal Description; Stroke-Segments-Mesh;
fLanguage
English
Publisher
ieee
Conference_Titel
Advanced Language Processing and Web Information Technology, 2008. ALPIT '08. International Conference on
Conference_Location
Dalian Liaoning
Print_ISBN
978-0-7695-3273-8
Type
conf
DOI
10.1109/ALPIT.2008.62
Filename
4584379
Link To Document