Title :
Structural analysis based stroke segmentation for Chinese characters
Author :
Lam, Josh H M ; Yam, Yeung
Author_Institution :
Dept. of Mech. & Autom. Eng., Chinese Univ. of Hong Kong, Hong Kong, China
Abstract :
This paper proposes a method for stroke segmentation to decompose Chinese characters to become individual strokes. The method first obtains the medial axes of any given Chinese character for structural analysis. A fast skeletonization method using Delaunay triangulation is applied to convert the character into triangular mesh. Based on the fundamental rules and formations of Chinese characters, the mesh is rearranged to remove the spurious branches existing in the character structure. Connections between the stroke branches are decided by those with the minimum differences in various factors, including gradients of the medial axes and boundaries, as well as the widths of the branches. Strokes are then generated and completed using Bezier curves. The work included experiments to examine the algorithm.
Keywords :
character recognition; image segmentation; mesh generation; Bezier curves; Chinese characters; Delaunay triangulation; character structure; fast skeletonization method; medial axes; stroke segmentation; structural analysis; triangular mesh; Acceleration; Filtering; Image analysis; Image reconstruction; Image segmentation; Information analysis; Joints; Shape; Turning;
Conference_Titel :
Decision and Control, 2009 held jointly with the 2009 28th Chinese Control Conference. CDC/CCC 2009. Proceedings of the 48th IEEE Conference on
Conference_Location :
Shanghai
Print_ISBN :
978-1-4244-3871-6
Electronic_ISBN :
0191-2216
DOI :
10.1109/CDC.2009.5400581